SPORTS MEDIA INVOLVEMENT VIA TEAM IDENTITY & ANTECEDENT MOTIVATIONS FOR THE PREDICTION OF TOTAL DAILY SPORTS MEDIA CONSUMPTION By Daniel Herman Krier A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Media and Information Studies – Doctor of Philosophy 2017 ABSTRACT SPORTS MEDIA INVOLVEMENT VIA TEAM IDENTITY & ANTECEDENT MOTIVATIONS FOR THE PREDICTION OF TOTAL DAILY SPORTS MEDIA CONSUMPTION By Daniel Herman Krier The present study investigates whether increases in sports team identity and sports media involvement correlate with increased amounts of daily sports media consumption. In addition, antecedents to involvement are incorporated into the research model to determine which discrete motivations show significant relationships with changes in involvement and time spent consuming. This project measured both team identity and sports media involvement on multidimensional levels to determine if influences by the constructs’ first-order facets would provide a richer source of explaining variance. These multidimensional predictor variables (six 3-item scales for team identity and three 3-items scales for sports media involvement) were tested in the model, as were similar unidimensional scales of team identity (4-items) and sports media involvement (5-items). It was discovered that the more parsimonious unidimensional scales explained more variance in the model predicting time spent consuming sports media, thus exposing the multidimensional scales as superfluous for a model such as the one tested here. From the perspective of fandom and fan identification, this study examined the attitudes and behaviors of adults who regularly watch or follow sports media in order to determine the strongest drivers of time spent consuming. Forms of media consumption included watching or following sports on television, the radio and podcasts, social networks and the Internet, and newspapers and magazines. These separate types of consumption were summed to produce the outcome variable of total time spent consuming sports media per day. The results of this project showed significant correlations between team identity, three antecedent motivations (vicarious achievement, escape from problems, and social media interaction), sports media involvement, and sports media consumption, while revealing partial and complete mediation. Additionally, gender differences regarding the motivational drivers of sports media involvement and time spent consuming were revealed by the predictive structural model. These findings illustrate the different ways in which male and female sports media consumers engage with sports media, thus providing a deeper understanding of what components drive sports consumption in conjunction with increased levels of sports team identity. The present study employed partial least squares structural equation modeling (PLSSEM) to test a path model consisting of a team identity instrument based on Social Identity Theory, antecedent motivations of involvement, sports media involvement, and finally, the effects of the preceding constructs on total time spent consuming sports media per day. The findings of this study are potentially helpful from both a communication and sports management perspective, as this project reveals the psychological drivers of today’s multimedia sports fans. Beyond the practical implications, this project builds on existing literature by combining related theories in a model that displays significant correlations and explains a large amount of variance of sports media involvement. The new 5-item unidimensional scale of sports media involvement discovered by this project can provide a useful tool for future research. It was found to be statistically reliable while illustrating convergent and discriminant validity, and most importantly was shown to be a significant behavioral predictor of sports media consumption. Copyright by DANIEL HERMAN KRIER 2017 ACKNOWLEDGEMENTS To my mother, Marion P. Krier, my father, Dr. Herman Krier, and my sister, Melissa. I thank my family for being so supportive of me my entire life. They have always encouraged me to strive to learn as much as I possibly could so that I would able to help others with the knowledge I have gained. I am extremely fortunate and appreciative of the wonderful opportunities afforded to me thanks to the hard work of my parents and the experience of my outstanding older sister. I love you all. To all the wonderful professors at Michigan State University who patiently helped teach me how to become a professional behavioral scientist, especially my advisor and mentor Dr. Stephen Lacy, and members of my dissertation committee, Dr. Manuel Chavez, Dr. Dawn Pysarchik, and Dr. Michael Stamm. I would also like to thank the chair of the MSU School of Journalism, Dr. Lucinda Davenport for welcoming me into the MSU family. Lastly, I would like to thank Nancy Ashley, whose invaluable knowledge and kind support helped me navigate the Media & Information Studies PhD Program. I am so grateful for the opportunity to work towards my doctoral degree over the last few years, and I could not be more proud to be a graduate of Michigan State University. This research was supported with funding from the College of Communication Arts & Sciences and the Graduate School of Michigan State University. v TABLE OF CONTENTS LIST OF TABLES ....................................................................................................................... viii LIST OF FIGURES ....................................................................................................................... ix CHAPTER 1 ................................................................................................................................... 1 INTRODUCTION ....................................................................................................................... 1 Female Sports Media Consumers and the Importance of Gender ............................................ 7 CHAPTER 2 ................................................................................................................................. 10 LITERATURE REVIEW AND HYPOTHESES ....................................................................... 10 Social Identity Theory and Team Identity .............................................................................. 10 Comparing Sports Fan Team Identity to Other Forms of Identity, Fandom, and Tribalism . 16 Involvement and Sports Media Involvement .......................................................................... 22 Motivational Antecedents of Involvement.............................................................................. 27 Linking Identity Level to Motivations to Consume, Involvement, and Sports Media Consumption ........................................................................................................................... 36 The Dependent Variable of Sports Media Consumption in Total Hours per Day .................. 38 Research Model, Hypotheses, and Research Questions ......................................................... 44 CHAPTER 3 ................................................................................................................................. 60 METHOD .................................................................................................................................. 60 Participants .............................................................................................................................. 60 Procedure ................................................................................................................................ 61 The Use of Partial Least Squares Structural Equation Modeling (PLS-SEM) ....................... 63 Data Analysis .......................................................................................................................... 65 CHAPTER 4 ................................................................................................................................. 67 RESULTS .................................................................................................................................. 67 Explanation of the Use of Unidimensional Scale Data Displayed in this Chapter ................. 67 Measurement Model ............................................................................................................... 68 Structural Model ..................................................................................................................... 75 CHAPTER 5 ............................................................................................................................... 104 DISCUSSION .......................................................................................................................... 104 Interpretation of Structural Model Results ........................................................................... 104 Gender Differences in the Consumption of Sports Media .................................................... 107 Managerial and Marketing Implications ............................................................................... 111 Social Identity Theory and Involvement Implications.......................................................... 112 Limitations and Future Research .......................................................................................... 116 vi APPENDIX ................................................................................................................................. 120 BIBLIOGRAPHY ....................................................................................................................... 130 vii LIST OF TABLES Table 1: Reliability of Constructs .............................................................................................. 34 Table 2: Operationalization of Constructs ............................................................................... 51 Table 3: Outer-Model Loadings and Cross-Loadings ............................................................. 70 Table 4: Descriptive Statistics and Reliability of Constructs.................................................. 71 Table 5: Variable Inflation Factor (VIF) – Multicollinearity Statistics................................. 72 Table 6: Correlations of Constructs & Discriminant Validity Assessment ........................... 73 Table 7: Heterotrait-Monotrait (HTMT) Matrix of Discriminant Validity .......................... 74 Table 8: PLS Structural Model Assessment ............................................................................. 93 Table 9: Mediating Effects Analysis .......................................................................................... 94 Table 10: Hypotheses Results .................................................................................................... 95 Table 11: First-Order Facets of Multidimensional Constructs .............................................. 98 Table 12: Gender Differences in Structural Model Relationships ....................................... 100 Table 13: Gender Descriptive Statistics .................................................................................. 100 Table 14: Unstandardized Consumption Coefficients Conversions into Minutes per Day 102 Table 15: Research Question Results ...................................................................................... 103 Table 16: Unidimensional Team Identity Items ..................................................................... 123 Table 17: Multidimensional Team Identity Items ................................................................. 124 Table 18: Antecedent Motivation Items .................................................................................. 125 Table 19: Sports Media Involvement Items............................................................................ 128 viii LIST OF FIGURES Figure 1: Proposed Model #1 of Multidimensional Constructs .............................................. 50 Figure 2: Proposed Model #2 of Unidimensional Constructs ................................................. 50 Figure 3: Complete Structural Model of Unidimensional Constructs ................................... 77 Figure 4: Team Identity Direct to Total Sports Media Consumption per Day ..................... 78 Figure 5: Team Identity Direct to Unidimensional Sports Media Involvement ................... 79 Figure 6: Sports Media Involvement Mediating Team Identity to Total Consumption ...... 80 Figure 7: Antecedent Motivations Mediating Team Identity to Sports Involvement .......... 82 Figure 8: Antecedent Motivations Mediating Team Identity to Total Consumption ........... 84 Figure 9: Direct Paths of Team Identity to Antecedent Motivations ..................................... 85 Figure 10: Direct Paths of Antecedent Motivations to Sports Media Involvement .............. 87 Figure 11: Direct Paths of Antecedent Motivations to Total Consumption .......................... 88 Figure 12: Complete Structural Model ..................................................................................... 90 ix CHAPTER 1 INTRODUCTION Sports media has become an integral part of American society in the 21st Century. It is part of many Americans’ lives on every single day of the year. As 70% of American adults regularly follow sports (Branscombe & Wann, 1991; Sporting News Media, 2014), fan interest has propelled sports media into a significant source of revenue for media companies across the country. These corporations aren’t merely involved in broadcasting live sporting events. Today, they are quenching fan interest by disseminating round-the-clock sports news, sports talk & analysis, written sports articles, statistical summaries, social media updates, and sports entertainment such as documentary shows and movies. As the Smartphone has become ubiquitous in recent years, sports fans can access live or recorded games and programming on demand, as well as articles, scores, recaps, and highlights at any time of the day. And with over 182.6 million Smartphone owners in the United States in 2015, with 220 million expected by 2018 (Statista 2015c), sports fans are increasingly interacting with one another and with athletes and those in the sports media industry via social media platforms such as Facebook, Snapchat, and Twitter (Hutchins, 2011). Simply put, not a day goes by without a large amount of newly produced sports media content available to be consumed by millions of people. It is very unlikely that this will change any time soon. The most familiar sports media conglomerate, ESPN – which launched in late 1979 – has helped grow the sports media world into a multibillion dollar industry both in America and worldwide (Statista, 2016b). The revenue generated by ESPN is a major reason its parent corporation, the Walt Disney Company, regularly boasts quarterly net income levels surpassing one billion dollars (Statista, 2016f). According to Forbes, ESPN was the most valuable media 1 brand in the world in 2015, worth $16.2 billion dollars, almost $5 billion higher than the secondplace brand, Fox (Forbes, 2015; Statista, 2016d). There is little doubting that sports media is an essential part of the economies of both the United States and the rest of the globe. Even as sports media options have become more diversified in recent years with the abundance of new technology (Sporting News Media, 2014), ESPN’s television viewership has held steady for the last eight years with a consistent average of 66 million individuals watching the network at least once a week (Statista, 2016a). The channel’s parent company, Disney, laid off around 300 upper-level, high-salary employees in late 2015 in order to cut operating costs as the company looks to curtail overspending (SportingNews.com, 2015). Even with these cutbacks, the large amount of viewership reveals that television remains the most popular form of sports media and that it has not waned significantly with the addition of new media consumption options for fans to devour. According to Rentrak/Pivitol Research, ESPN has a monthly reach into 48% of American households (Statista, 2015b), making it one of the sixthmost popular network in the United States in 2015 after CBS, NBC, ABC, Fox, and Univision (TV Insider, 2016). Sporting News Media’s 2014 report on US Sports Media Consumption pointed out that 70% of the adult population in America – over 168 million individuals – follow sports daily, with an even distribution among all age demographics (Sporting News Media, 2014). 55% of these sports fans are male, 96% of sports fans consume their sports on television, 68% follow sports online on a daily basis, and 35% consume sports via social networking platforms (Sporting News Media, 2014). The last number is surely bolstered by the finding that 42% of fans consume content via a mobile device such as a Smartphone or Tablet. From 2006 to 2015, 2 the business of sports media rights in North America nearly doubled from $8.49 billion to $16.37 billion (Statista, 2016e). That value is expected to eclipse $20.6 billion by the year 2020. What these numbers tells us is that sports media is undoubtedly a major part of American society and the US economy. These statistics are consistent over the past decade, and the already-record totals are going to grow larger due to billion-dollar contracts that are in place spanning years into the future. In late April 2016, the Big Ten Conference signed a deal with Fox Entertainment Group to sell half of their rights for sports media for a value of $250 million a year (Sports Business Daily, 2016). This lucrative contract will enable the conference, which consists of just 14 teams, to reap many more millions from media rights deals with ESPN and CBS for football and basketball broadcasts and streaming. With thousands of sports teams competing in many types of sports that are disseminated and covered by the media day in and day out, amateur and professional alike, it is no surprise to hear that global sports media revenue, of which 34% is from North America, totals out to more than $35 billion annually (PricewaterhouseCoopers, 2015; Statista, 2014a). Beyond financial statistics, sports and sports media are interwoven into American culture. Each fall, 49% of all adult Americans report watching and following NFL football (Sporting News Media, 2014). This fan interest manifests in both psychological and behavioral connections with the teams for which these involved individuals cheer. In Seattle, for example, the NFL’s Seahawks’ fan base collectively refers to itself as the “12th Man” (M. T. Brown, Zuefle, & Batista, 2007), which speaks to the unity of sports team identity embodying a single cohesive mindset. Behaviorally, fans regularly purchase team-related merchandise, which has seen US domestic sales between $7 and $8 billion dollars annually since 2008 (Statista, 2015a). Furthermore, the nature of sports media helps merge the psychological and behavioral by 3 bringing individuals together in restaurants and sports bars like Buffalo Wild Wings to watch games of their favorite team with fellow fans, even in cities far away from home where rival teams are based (A. C. Osborne & Coombs, 2013). Due to the social and economic influences of the sports media industry on American culture, this study will examine the relationships between team identity, motivations to consume sports media, and sports media involvement with changes in time spent consuming sports media. A model consisting of psychometrically sound scales with internal consistency and construct validity is used to empirically test the connections between attitudes about sports teams and sports media with the behavior of consuming sports media (Beaton, Funk, Ridinger, & Jordan, 2011; Funk, Ridinger, & Moorman, 2004; Heere et al., 2011; Wann & Branscombe, 1993). It is beneficial to develop a deeper understanding of the associated factors driving the consumption of sports media. These insights are valuable from two important standpoints. First, from a marketing perspective, the motivational antecedents to involvement will reveal the significant factors that correlate with sports media involvement, a concept that recent literature points out needs to be more thoroughly tested and understood (Beaton et al., 2011). Uncovering the relationships between team identity, antecedent motivations, and sports media involvement will enable sports marketing managers to craft specific media messages to potential consumers that could greatly increase time spent consuming sports media annually. In addition, by measuring team identity and sports media involvement as second-order multidimensional constructs (Beaton et al., 2011; Heere et al., 2011; Wetzels, Odekerken-Schröder, & van Oppen, 2009), this project will seek to uncover the varying degrees of importance of the six first-order facets of sports fan identification (Heere & James, 2007b) and the three first-order facets of sports involvement (Beaton et al., 2011). 4 Secondly, from a communication research standpoint, this project will provide insight on the social and psychological impact of sports media consumption. Research has noted several positive aspects of identifying with a sports team (Branscombe & Wann, 1991). Such characteristics of fan identification include enhanced levels of self-esteem, reduced feelings of depression, and a lowered sense of alienation from the rest of society. Additionally, a personal connection with a sports team provides an enhanced sense of belongingness to the community at large and the outlook that sports fans are involved in something greater than themselves. The current study will explore motivations that arise from this feeling of team identity such as the sense of being able to escape one’s daily problems due to media consumption as well as experiencing vicarious achievement by seeing teams and favorite players win at competitive events. Furthermore, other research points out additional positive benefits to society from participating in sports and being a sports spectator (Eime, Young, Harvey, Charity, & Payne, 2013; Theokas, 2009). This is especially true for children who are in developmental stages of social growth who are involved in extracurricular activities like sports participation (Zaff, Moore, Papillo, & Williams, 2003; Zarrett et al., 2009). Not only does sports participation and spectatorship provide positive outcomes for young people like heightening communication skills, dealing with adversity, and finding a balance between work and leisure, it can teach children about comradery, respecting authority, and developing dedication to achieving an overall goal. Identifying with a sports team can help individuals learn about the importance of working with others to achieve a common goal while seeing firsthand the resilience necessary to find success in life. Overall, this study will increase insight on the underlying sociological drivers of the psychological connections related to a sense of team identity, thus providing a better explanation 5 of what it means to be a sports fan in the 21st Century. The various factors of team identity, antecedent motivations of involvement, and sports media involvement have been shown in prior research to have indirect and direct effects on one another, as they have each been tested separately relating to behaviors connected with sports consumption (Funk, Mahony, & Ridinger, 2002; Funk et al., 2004; Heere & James, 2007b; Heere et al., 2011; Trail, Fink, & Anderson, 2003; Wann & Branscombe, 1993). As such, these variables are expected to significantly correlate with one another when tested in a model incorporating each construct. Never before has a study linked social identity theory with sports media involvement, nor has one focused on the relationship of those two constructs with time spent consuming sports media. By simultaneously testing team identity, antecedent motivations of involvement, and sports media involvement – all variables that consist of psychological connections with teams and sports media – the simulation will produce a blueprint of the mental drivers of modern sports media consumption for today’s sports fans. Additionally, this project will provide deeper insight into sports media consumption by extending and replicating a prior study that discovered significant relationships between these variables (Krier, 2016). It is expected that this research project will reveal significant findings regarding the discriminant constructs that play important roles in assessing what constitutes today’s sports media consuming fan, as well as any possible gender differences. Discovering the important components that drive sports media consumption, and the magnitude of each element, is imperative for sports marketing managers and communication scholars to understand. It will shed light on who sports fans are in the 21st Century and what predicts their consumption of sports media. This project will enable communication and marketing scholars alike to better understand, predict and control the relationships between sports fan identity, antecedent motivations of involvement, sports media 6 involvement and the daily consumption of sports media. It will provide a comprehensive picture of the thoughts and behaviors of the modern sports media consumer. Female Sports Media Consumers and the Importance of Gender Sports media is a billion-dollar industry around the world that continues to grow each year (Statista, 2016b). Gender plays a significant role in the business of sports media due to the large proportion of female sports fans and consumers, with these numbers increasing year by year as well. For instance, Scarborough Research reported that adult women in the United States who classified themselves as sports fans increased from 29% to 58% from 1998 to 2002 and notes that women were the buyers of 70% of the National Football League’s officially licensed products (Bush, Bush, Clark, & Bush, 2005). Furthermore, women spend more than $7 billion each year on consumer goods in the United States, and they influence 85% of all purchase decisions, including a majority of products made for men (Staurowsky, 2016). A study by Adweek reported that females make up 46% of all NFL fans, with 63% of women over the age of 12 identifying as sports fans (Adweek, 2014). Women embody a substantial segment of the sports media market in terms of meaning and importance when it comes to marketing decisions by media managers. This is why media entities like espnW, an offshoot of ESPN covering only women’s athletics and specifically geared towards female sports media consumers, have grown into a vital part of sports journalism in recent years (Wolter, 2015). Therefore, it is of great importance to pinpoint the drivers of female sports media consumption and to compare that to the consumption of males to identify any significant gender differences (Billings, Butterworth, & Truman, 2015). If these two groups of sports media consumers have separate motivating factors driving consumption, this signifies that there are two separate cultures that need to be approached differently by marketing messages 7 Since there is such a high female interest in sports media for both men’s and women’s athletics, crafting a message that is gender-specific is something that sports media organizations need to develop in order to reach this large, previously-ignored culture (Forbes, 2014). Accordingly, it would be valuable to determine the motivations behind women’s consumption of sports media, and how these motivations predict changes in total sports media consumption per day. This is especially true if women differ from men. Females are too often taken lightly in their ability to affect industry of sports media, especially being that they have such a distinct ability to influence other women and young girls via social media and electronic word-of-mouth (Dix, Phau, & Pougnet, 2010). Beyond sports, women and men represent different cultures with respect to their consumption of entertainment and emotional reactions to media. Consequently, various forms of media are frequently based on gender stereotypes (Oliver, Weaver III, & Sargent, 2000), which cause different media experiences for men and women (Hartmann & Klimmt, 2006). Research notes that women experience mediated sporting events differently than men (Gantz & Wenner, 1991), that women’s athletics are covered differently by sports media (Creedon, 1998; Wolter, 2015), and that women differ from men in the ways in which they consume sports media (Staurowsky, 2016; Wenner & Gantz, 1998). Finally, gender has been used to examine a wide range of marketing applications and consumer behavior applications like emotional responses to mediated messages (Orth, Malkewitz, & Bee, 2015), the marketing of Smartphones (Persaud & Azhar, 2012), customer loyalty (Ndubisi, 2006), and the specific motivations and preferences of sports spectators (M. J. Robinson & Trail, 2005). As we approach the close of the second decade of the 2000’s, it is vital to determine the motivational drivers of female sports media consumers to define cultural differences in sports 8 media consumption based on gender. Being able to explain the motivations of the different genders of sports fans will be useful from not only a cultural perspective, but potentially from a marketing and sports management standpoint as well. 9 CHAPTER 2 LITERATURE REVIEW AND HYPOTHESES Social Identity Theory and Team Identity Social Identity Theory (SIT) proposes that an individual’s self-concept is a consequence of one’s knowledge of group membership where emotional significance is attached to being part of that group (Tajfel, 1978). The theory suggests that individuals categorize themselves into various social alignments in order to facilitate self-definition within their own social environment (Ashforth & Mael, 1989; Hogg & Hains, 1996; Hogg & Reid, 2006; Tajfel, 1974), thus helping to answer the question: “Who am I?” It enables individuals to explain to themselves and others who they are when they speak of “we” as a collective identity (Brewer & Gardner, 1996). According to SIT, by seeking out and maintaining relationships with favorable groups, people advance their overall self-esteem to achieve and maintain a positive self and social identity. An important aspect of SIT is that identification with a group or organization does not have to be physical in nature. As long some type of emotional attachment exists, individuals need only to psychologically feel they are a part of a group in order to acquire a sense of belonging (Ashmore, Deaux, & McLaughlin-Volpe, 2004). A sociological viewpoint presents self-identity as a reflection of the environment surrounding individuals and the narrative that helps people answer essential questions about their personal existence (Giddens, 1991). The modern world provides many different cues for individuals to process in order to assess the consequences of their actions and decisions. This information helps aid uncertainty reduction and risk aversion, thus shaping a person’s identity as someone who is living according to a personally structured lifestyle. In other words, self-identity is a result of answers to questions about what to do and how to act for each individual (known as 10 reflexive monitoring), thus shaping the knowledge behind self-understanding. Motivations behind why people act the ways in which they do are the source behind actions, and are conjured by the need to establish a self-identity. Mediated forms of communication can help to structure experience and aid self-definition by presenting an organized source of information for people to consume. When people can experience the thoughts and actions of others with similar physical and psychological characteristics via mediated communication (books, television shows, movies, online, etc.), it provides a unifying factor thanks to modernity in an ever-changing world (Giddens, 1991). Another characteristic of SIT is the in-group versus out-group dynamic (Karasawa, 1991). As individuals associate psychologically with favorable groups, it indicates that they are not only defining who they are, but rather who they are not (Branscombe, Wann, Noel, & Coleman, 1993; Tajfel & Turner, 1979). Such feelings result in positive assessments of one’s ingroup while producing biased, negative valuations of the out-group (Eckel & Grossman, 2005). These emotions are regularly found in the minds of sports fans that identify with a team, while opposing teams are thought of in negative terms and as threats to their group’s integrity (Wann & Grieve, 2005). Interestingly, the dynamic does not need to be based on anything tangible or even real. This is because once self-categorization sets off the mechanisms by which individuals knowingly belong to an in-group, the mere knowledge of belonging to a group results in outgroup bias and discrimination (Grieve & Hogg, 1999). This phenomenon, tested with the minimal group paradigm, shows that individuals naturally regard their own groups as more positive than those with which they feel they have no psychological connection (Tajfel, 1970; Tajfel, Billig, Bundy, & Flament, 1971). 11 The results of identifying as part of the in-group with a sports team correlates with other psychological feelings towards sports media, the motivations for consuming sports media and the behavior of time spent consuming (Krier, 2016). Due to the psychological connection with a favorite sports in-group, fellow fans of the same team can use observable symbols and cues as a way to guide reliable patterns of behavior (Trail et al., 2003). This can result in such actions as watching more sports media of a favorite team and buying more products seen advertised via that team’s various media (Hunt, Bristol, & Bashaw, 1999). Other acts of consumerism that manifest from increased levels of team identity include such behaviors as regularly purchasing team clothing and merchandise, as well as the loyal consumption of associated team media (Funk & James, 2006). Additionally, SIT helps to support the explanation of sports team affiliation and the notion of high-identity fans (Wann, Dolan, McGeorge, & Allison, 1994). While sports fans have been labeled “die-hard” or “fair-weather” based on their reactions to sporting event outcomes (Wann & Branscombe, 1990), the notion of team identity level allows for the measurement of this construct as a continuous variable. Specifically, team identification refers to the extent to which a fan senses a psychological connection to a team, such that there is a self-relevance connected with and the team’s performances on the field of play and off (Wann, Raney, & Bryant, 2006). In this regard, researchers have found a positive relationship between self-esteem and identification with a sports team where the perceptions of high-identity fans differ from others, especially concerning one’s perceived effect on sports outcomes (Wann et al., 1994). Identification with a sports team was first measured with instruments such as the Sports Spectator Identification Scale or SSIS (Wann & Branscombe, 1993) and the Psychological Commitment with the Team scale or PCT (Mahony, Madrigal, & Howard, 2000). The SSIS 12 defines team identity as a person’s level of attachment to a specific team or organization (Wann & Branscombe, 1993). Both the SSIS and PCT scales have been deemed valid for interpreting team identification level (Wann & Pierce, 2003), although the PCT was developed for segmenting sports fans based on a continuum of loyalty. The SSIS and PCT are similar in that they measure team identity on a unidimensional level (Wann, 2006). The SSIS has been shown to be a valid measurement of team identification in a model that significantly links increased team identity level to higher motivations to consume sports media and total sports media consumption (Krier, 2016). As one’s degree of social identity is said to determine the power to which that identity will influence associated behaviors (Solomon, 1983), a unidimensional scale of team identity based on the SSIS is tested in the current to determine if it is a significant predictor of total sports media consumed per day. Team identity has also been measured multidimensionally as a second-order construct in order to help study the effect of the identity on consumption behavior (Heere & James, 2007b). Multidimensional variables consist of one or more first-order facets, which are separate variables that are each measured with valid and reliable scales that combine to contribute to the calculation of the second-order variable (Edwards, 2001; Wetzels et al., 2009). As a multidimensional measure, team identity was determined to consist of six first-order facets: public evaluation, private evaluation, interconnection of self with the team, sense of interdependence with the team, behavioral involvement, and cognitive awareness (Ashmore et al., 2004; Heere & James, 2007b; Heere et al., 2011). Public and private evaluation assess degrees of collective self-esteem regarding how much pride is connected to the performance and perception of a favorite team (Luhtanen & Crocker, 1992). Interconnection of self with the team and sense of interdependence are both subscales of attachment measuring emotional connections with the team and its 13 performance (F. A. Mael & Tetrick, 1992). Behavioral involvement assesses participation activities of fans that expresses identity, such as attending games, wearing team merchandise, and discussing the team with other fans (Ashmore et al., 2004). Regarding nomenclature, it should be noted that the first-order facet called behavioral involvement from the Heere et al., 2011 study does not directly relate to sports media involvement, which is tested as a latent variable in the current study’s structural equation model. Sports involvement, from which sports media involvement is based, is a multidimensional construct comprised of three facets, none of which measure attending games, wearing or displaying team merchandise, or discussing sports with others (Beaton et al., 2011). The sixth and final first-order facet of multidimensional team identity is cognitive awareness, which measures knowledge of the history and traditions of one’s favorite sports team and fan base (Heere & James, 2007b). The individual first-order facets of the multidimensional team identity scale were developed from previously operationalized items that were tested for internal consistency and validity (Gurin & Townsend, 1986; Luhtanen & Crocker, 1992; F. A. Mael & Tetrick, 1992; Phinney, 1992). Due to the content validity of the multidimensional team identity scale, the items from Here et al., 2011 is used in the current study to assess the overall level of identity and degrees of importance for each facet. Furthermore, this research project will conduct a comparison of the two types of measures – unidimensional (Wann & Branscombe, 1993) vs. multidimensional team identity (Heere & James, 2007b; Heere et al., 2011) – with two separate models to determine if any additional variance in total sports media consumption is attained by employing the multidimensional scale of team identity. The segmentation of sports fans based on identity level has been examined in previous literature using the Psychological Continuum Model or PCM, which categorizes the four stages 14 of a fan’s connection with the sports team (Funk & James, 2001). Moving from low to high, the PCM places fans into the segments of Awareness, Attraction, Attachment, Allegiance regarding psychological relationships with a team. Authors of that study later refined the PCM to explain that there are expected outcomes for each level of relationship with the team and that moving from Attraction to Allegiance is mediated by levels of Attachment, making the emotional connection to one’s team a dynamic process (Funk & James, 2006). As mentioned earlier, level of attachment is at the core of defining what constitutes sports team identification (Wann & Branscombe, 1993). While these categories for assessing identity level are important to evaluating sports fans from a developmental level (Lock, Taylor, Funk, & Darcy, 2012), the antecedents to team identity will not be measured in this study. That said, factors that may lead to a feeling of group identity with a sports team or organization include such characteristics as group performance and group attractiveness, with strengths of each influence based on the success of the sports team (Fisher & Wakefield, 1998). For instance, a losing team might have individual players whose skill at performing their athletic duties is worth attracting fans. From a theoretical standpoint, SIT has been used specifically to examine possible motivations to identify with a sports team. Those studies remarked that uncertainty reduction and the desire to advance self-esteem are both likely antecedents to team identification (Dimmock & Gucciardi, 2008; Hogg & Grieve, 1999). Other studies point out antecedents to organizational identity such as group distinctiveness, prestige, and competition with other organizations in the form of out-group salience (F. Mael & Ashforth, 1992). Those authors noted that factors customarily accompanying group formation – such as physical similarity, liking, physical proximity, shared goals, common history, and interpersonal interaction – can possibly affect the degree to which someone associates with an 15 organization, though they are not necessary for psychological connections to occur (Ashforth & Mael, 1989). Conversely, some research specifically points out that liking other group members has been shown to be a consequence of a sense of group identity, not an antecedent (Brewer & Gardner, 1996; Hogg & Turner, 1985). In conclusion, sports team identity level is an important construct to investigate based on Social Identity Theory because it has been shown to strongly affect media consumption (Heere et al., 2011), and it also represents the first step in the mechanism leading to a greater sense of involvement (Underwood, Bond, & Baer, 2001). It has been stated that involvement is a reflection of one’s sense of identity (Traylor & Joseph, 1984). By establishing the significant associations between team identity and sports media involvement, as well as motivational origins of such involvement (Funk et al., 2004), a valid comprehensive assessment of the factors predicting sports media consumption can be better judged and evaluated. Comparing Sports Fan Team Identity to Other Forms of Identity, Fandom, and Tribalism As the significance of team identity regarding sports fans was examined in the previous section, it is important to compare this type of psychological connection with other similar forms of identity. These other manifestations include such classifications as national identity, religious identity, ethnic identity, sexual identity, and gender identity, which have all been described in some studies as a collective identity (Ashmore et al., 2004). Other typologies have explained group formation and identity in terms of fandom and tribalism (Dionísio, Leal, & Moutinho, 2008; Melnick & Wann, 2011). Consumers have also been linked with brands in a tribal fashion regarding a collective devotion to a particular product or company (Sierra, Badrinarayanan, & Taute, 2015). Additionally, sports organizations can be seen as tribal brands due to these fervent 16 types of psychological connections between a team and its fans (Gladden & Funk, 2002; Meir & Scott, 2007). Sports fans have been characterized as individuals who exhibit durable features of identification that include cognitive, affective, and behavioral connections with a team (DietzUler, Harrick, End, & Jacquemotte, 2000; Lock, Darcy, & Taylor, 2009). The team and its associated community are seen as representations of family for a sports fan, which corresponds with more predictable feelings and behaviors due to desirable qualities of the relationship (Branscombe & Wann, 1991). For instance, high-identity fans of a team have been shown consistent behaviors like attending games more frequently and displaying team merchandise (Cialdini et al., 1976; Murrell & Dietz, 1992). From an emotional or affective standpoint, high identity sports fans have been found to be more satisfied with their teams victories, as well as an overall higher enjoyment with sporting events (Madrigal, 1995; Wann & Schrader, 1997). Cognitively, sports fans that highly identify with a team are more knowledgeable about their favorite organizations than non-fans and are able to communicate with other group members using insider jargon (Wann et al., 1997). These types of sports fans have been shown to gain more positive social-psychological health benefits due to their identification with the team (Wann, 2006). Lastly, high-identity sports fans feel that they have a real influence on the outcome of sporting events than low-identity individuals (Wann & Dolan, 1994). Other behaviors that manifest from increased identity include sports fan rituals. Rituals are actions or behaviors performed by groups in a consistent manner to serve the symbolic purpose of community togetherness (Billings et al., 2015). These rituals can represent important rites of passage within any group lineage providing an increased sense of community. Examples 17 include tailgating outside of sports venues before and after games, face painting and dressing in team merchandise, singing in unison during events, and more extensive sports fan lifestyle rituals like NASCAR weekends for racing fanatics. Such traditions pass from one generation to the next and adapt to the changing world and cultural environment as time goes on. There are many other examples regarding sports game attendance and media consumption habits such as going to events with parents and sitting in season ticket seats that have been in the family for 30 years or watching the Detroit Lions every Thanksgiving. The result of rituals for sports fans is a heightened sense of emotional importance and an increased value of being a member of the group (Billings et al., 2015). Concerning attitudes related to sports teams, there are differences in behavioral responses to affective and cognitive influences regarding team identification. Affective responses are said to have a wide range of effects on evaluative behavior and social interaction with sports while cognitive responses relate to expectancies of future sports outcomes (Smith, 2006). Measurements of affective or emotional factors have contributed to assessments of brand attitudes regarding sports teams (Bauer, Stokburger-Sauer, & Exler, 2008). Research about cognitive aspects of identity has stated that measurements of cognitive awareness, or knowledge of group membership, is the first step in defining team identity level (Lock et al., 2012). Other research has proposed that cognitive factors precede affective influences and that those two aspects of attitude allow for the prediction of behaviors towards the sports team such as a higher amounts of sports consumed and more frequent game attendance (Tsiotsou, 2013). Fandom has been defined as a form of expressing oneself in terms of a relationship to a shared community with intimate attachments and allegiances (J. A. Brown, 1997). It differs from team identity in that fandom can manifest from psychological relationships with non-sports 18 entities such as comic books, television characters, or online gaming portals. Fandom manifests in shared preferences, beliefs, and activities within a culture, and like social identity, there is the delineation between “us” and “them” for those within the community (Alvermann & Hagood, 2000). Also relating to a sense of common identity, the aspect of allegiance involving fandom serves as the backbone of a psychological relationship with heightened sense of emotional importance and value results from being a member a group (Hirt, Zillmann, Erickson, & Kennedy, 1992). Sports fandom has been examined in existing literature, with the prevailing outcome that there is not one clear definition of exactly what constitutes the concept (A. C. Osborne & Coombs, 2013; Williams, 2007). Rather, the main difference between individuals regarding sports fandom comes from what sports fans are not; sports fans are not mere passive spectators with no emotional connection to outcome of the game. Sports fandom, like team identity, has been shown to be contingent on the psychological connections with the team and the shared feelings with other group members (A. C. Osborne & Coombs, 2013; Williams, 2007), regardless of the place from which the person might be cheering (Kraszewski, 2008). Several studies have inspected the differing levels of sports fandom in an effort to classify and segment individuals into varying fan typologies (Hunt et al., 1999; Stewart, Smith, & Nicholson, 2003; Tapp & Clowes, 2002). While these nominal groupings differ in each study, each of them orders fandom from low to high in terms of team identity level. This illustrates that there is a degree of fandom that is always present when considering sports team identity level. As was noted in the previous paragraph, defining sports fans and fandom has consistently been accomplished via comparisons to other forms of social identity (A. C. Osborne & Coombs, 2013). 19 One type of identity experienced at some point by nearly everyone in life is religious identity. Participation with religion has been shown to be positively correlated with feelings of optimism and subjective well-being (Salsman, Brown, Brechting, & Carlson, 2005; Witter, Stock, Okun, & Haring, 1985), which are similar traits associated with identity discussed by SIT (Ashforth & Mael, 1989). These positive social psychological benefits have been investigated in a sports setting as well with consideration to religious behavior (Wann, 2006). The study concluded that as in religion, the relationship between team identity and well-being might be a function of the social opportunities to interact with others from the same religion, or sports team. In other words, both religion and sports teams facilitate the feeling of individuals being part of a greater whole which helps to define the self and allow for positive self-categorization. Similarly, concepts of ethnic identity and gender identity also adhere to the correspondent psychological connections that individuals feel they share with a collective whole. Studies on gender identity pointed out that perceived similarity one’s group along with how outsiders treat group members – known as “common fate” – played significant roles in helping to structure an individual’s sense of self (Gurin & Townsend, 1986). The self-categorization aspect of SIT allows people to determine their own gender identity simply from an understanding of outward cultural definitions of being male or female (Schmader & Block, 2015; Wood & Eagly, 2015). Ethnic identity, which is not the same as outward labeling of ethnic groups, has also been defined with regard to how individuals classify themselves. In order to facilitate self-definition of one’s ethnic identity, people take into account the cultural aspects of their group using cues such as group values, behaviors, language, and an understanding of the group’s history (Phinney, 1990, 1992). In each case, it is up to the individual to decide what group promotes an identity that advances his or her own self-categorization and self-esteem (Tajfel, 1982). 20 In regards to comparing and contrasting types of identity, it is important to state that identity is primarily a psychological function of what a person thinks he or she is (Ashmore et al., 2004), regardless of outside appearances (Brewer & Gardner, 1996). While it is true that some types of physiological traits like gender or race or ethnicity are categories not chosen by individuals at birth, measurement of these types of identities in research asks people what they “consider” themselves to be (Hoffman, 2006; Phinney, 1990; Wood & Eagly, 2015). For instance, this does not mean a person who was born as a male cannot identify as a female, or that a person with one type of pigmentation in his or her skin cannot identify with another race. Rather, personality traits dealing with gender have been shown to be drawn from stereotypes of how people think males or females are likely to behave (Wood & Eagly, 2015). Gender selfdefinition is a reflection of how strongly a person considers their actions to be masculine or feminine (Hoffman, 2006). Similarly, ethnic and racial identities are reflections of individuals’ understanding and interpretations of what they feel their own ethnicity to be (Phinney, 1996). While it would seem that one would have more choice as to what sports team he or she identifies with, research shows that human identity is a function of what people feel they are more than what others define them to be. Therefore, choosing to cheer for a team that is based in a different city than the one sports fans currently live in (Kraszewski, 2008) is similar to the other types of identities individuals can consider themselves to be. Finally, existing research has recognized other types of identity which result in cultures of consumption resulting from the group identity perceived by its members. For instance, motorcycle riders who owned Harley Davidsons were studied in attempt to discover the organized consumption habits of those who felt like a member (Schouten & McAlexander, 1995). The results revealed that that not only did members feel they were a singular voice when 21 hearing other motorcycle engines roar with theirs in unison, isolated Harley owners exhibited similar purchasing patterns and conformist consumption practices to reflect their sense of group identity. It was concluded that Harley Davidson owners’ identity was inseparably linked to their purchase and consumption habits. One of the researchers of that study had previously conducted ethnographic interviews of individuals who had undergone facial surgery in hopes of transitioning to a literal reconstructed identity (Schouten, 1991). An unforgettable example discussed in that article spoke of a woman who underwent facial surgery in order to erase visible lines brought upon her due to the grief of being a widow. The facial reconstruction allowed the individual to again feel like she was part of the group to which she used to belong and no longer had to identify with what she felt was causing her emotional torment. Therein lies the power of social identity: we are free to identify however we like with whomever we choose and in whichever way increases our self-esteem, our positive self-definition and our self-categorization (Bergami & Bagozzi, 2000). In conclusion, no matter which type of identity an individual chooses for him or herself, the decision lies ultimately with every person regardless of external factors or physical characteristics. Better yet, people can define multiple types of identities such as the group categories discussed in this chapter provided the sense of identity positively advances selfcategorization and self-esteem (Luhtanen & Crocker, 1992). Involvement and Sports Media Involvement Initially, the construct of involvement was explained as a psychological state that reflected the personal relevance and importance of an issue (Sherif, Sherif, & Nebergall, 1965). From that point forward, studies on involvement have defined and operationalized the concept in substantially different ways without a solid agreed-upon consensus or precise definition (Beaton 22 & Funk, 2008; Costley, 1988; Jain & Srinivasan, 1990; Zaichkowsky, 1986). There is a wide array of involvement concepts including leisure involvement (Havitz & Dimanche, 1997, 1999; Iwasaki & Havitz, 2004), ego involvement (Miller, 1976; Selin & Howard, 1988), job involvement (Lodahl & Kejner, 1965), task involvement (Skaalvik, Valåns, & Sletta, 1994), brand and product involvement (Bloch & Bruce, 1984; Knox & Walker, 2003; McQuarrie & Munson, 1992), and behavioral involvement (Heere et al., 2011; Roccas, Sagiv, Schwartz, Halevy, & Eidelson, 2008), which also happens to be a first-order facet of multidimensional sports team identity (Heere & James, 2007a, 2007b). Despite this lack of consensus and multitude of sub-categories, similarities among each type of involvement pertain to how important, relevant, or central an object or activity is to a person’s life (Johnson & Eagly, 1989). Involvement has also been defined as a psychological state of arousal or motivation evoked by some type of stimulus of interest or situation that propels an individual to behave in a more dedicated manner than previously existed (Rothschild, 1984). Furthermore, that study’s definition cited work that suggested involvement is a continuous variable able to be operationalized and measured (Antil, 1984). The motivated state of mind accompanying involvement is relevant in many studies inspecting consumer behavior (Laurent & Kapferer, 1985; Roy & Goss, 2007; Traylor & Joseph, 1984). This is because an individual with an amplified interest in an activity such as sports media consumption will be likely possess a greater intention to consume, which in turn could lead to increased time consuming sports media (Funk et al., 2004; Tsiotsou, 2013). Additionally, involvement has been shown to be a mediating variable between attitudes and behaviors regarding consumer behavior (Mitchell, 1979; Ray, 1979; Zaichkowsky, 1985). In this research project, the multidimensional construct of sports involvement (Beaton et al., 2011) present in the 23 measurement model is predicted to serve as a mediator linking team identity to sports media consumption. The idea of involvement being a motivated psychological state is important to this study because it provides a link to one’s sense of identity through the feelings of a personal connection with consuming a service or product (Zaichkowsky, 1986). Sports teams and athletes have been shown to have direct connections with individuals and communities such that the more involved a person is, the more one can be expected to consume associated sports media of that team (Walker & Kent, 2009). The goal of the sports marketing branches of an athletic organization or sports media company is to promote and support a more involved fan base that identifies socially with the team and athletes (Fullerton & Merz, 2008). In this manner, team identity allows individuals to have an array of dependable patterns in their lives due to a psychological connection and increased motivation to consume the media of a favorite sports team (Heere & James, 2007b; Underwood et al., 2001). The more involvement a person feels, the more attentional capacity he or she will devote to a message source, which is an important feature of consumer behavior, especially in relation to consuming sports media (Greenwald & Leavitt, 1984; Mittal & Lee, 1989). Like social identity, involvement is solely up to the individual’s personal preferences and feelings. If a media organization can foster more of sense of team identity and involvement, it is logical to conclude that the fan base will be more predictable and consistent consumers. Existing research has noted that social identity with a group is an antecedent to attitude importance (Boninger, Krosnick, & Berent, 1995). Attitude has been defined as a psychological disposition that is expressed towards a certain object or activity with some degree of favor or disfavor (Eagly & Chaiken, 2007). As increased involvement has been shown to affect the extent 24 to which someone is willing to change his or her attitude towards a product or service (Barki & Hartwick, 1994; Kokkinaki & Lunt, 1997; Sherif et al., 1965), it can be said that one’s identity level could possibly determine the level of one’s attitude acceptance or rejection, or involvement. Thus, the link between identity and involvement is established. Taking this one step further, research has stated that a positive attitude correlates with increased consumption intention in conjunction with increased team identification (Madrigal, 2001). Due to these theoretical associations, the predictive structural equation model in the current study will explore the correlations between team identity level, sports media involvement, and the amount of time spent consuming sports media. Additional direct links between social identity and involvement have been discussed in prior research. One study pointed out that an increased level of involvement strongly correlated with higher levels of team identity (Fisher & Wakefield, 1998). Another detailed the sports marketplace by exploring the relationship between involvement and social identity (Underwood et al., 2001). That study determined that a sports organization could foster increased customer patronage by creating more personal relevance, or involvement, by highlighting a team’s historic traditions and monolithic stadium facilities – reminders of long-term team identification. Sports were determined to lie at the high involvement end of the identification continuum. Measurement of the various involvement constructs mentioned at the start of this chapter have been as distinct as the types of involvement that have been analyzed in academic research. For instance, one of the earliest and most cited scales of consumer involvement were tested with the Personal Involvement Inventory or PII (Zaichkowsky, 1985). The items in the PII asked about feelings of personal importance for products such as instant coffee, laundry detergent, and color televisions. The PII has been modified to test other consumer products (McQuarrie & 25 Munson, 1987), the effects of advertising (Zaichkowsky, 1994), and sports team involvement (Tsiotsou, 2013). Another early influential scale of involvement was deemed the Involvement Profile or IP Scale (Laurent & Kapferer, 1985). The IP Scale was designed to offer a more comprehensive profile of a person’s involvement in order to better understand the consequences on consumer behavior. In order to assess the involvement profiles of sports spectators, researchers used the IP scale to measure involvement as a multidimensional construct consisting of four facets (Kerstetter & Koovich, 1997). The four first-order dimensions of involvement, based on the original IP Scale, were determined to be Perceived Importance, Perceived Risk, Symbolic Meaning or Sign, and Hedonic Value or Pleasure. Similarly, the construct of leisure involvement was determined to consist of three facets of enduring involvement – Attraction, Symbolic Meaning or Sign, and Centrality to Lifestyle – and two facets of risk involvement – Risk Probability and Risk Consequence (Havitz & Dimanche, 1997, 1999; Iwasaki & Havitz, 2004). Lastly, a study on the origins of involvement with professional sports teams proposed five facets of involvement: Perceived Importance, Risk Importance, Risk Probability, Symbolic Meaning or Sign, and Hedonic Value or Pleasure (Funk et al., 2004). The current research project will measure involvement based on a more recent study that employed a more parsimonious construct based on to the IP Scale to assess levels of sports involvement (Beaton et al., 2011). As that study pointed out the in the substantial ways in which involvement has been conceptualized and operationalized, the authors wanted to test a valid measure of one’s involvement that would be of utility in the context of sports management. To accomplish their goals, sports involvement was determined to be a multidimensional construct comprised of three first-order facets based on past measures of leisure involvement and enduring involvement (Kyle & Mowen, 2005; McIntyre, 1989). The first facet, Hedonic Value, refers to 26 the pleasure or enjoyment felt from doing an activity, the second, Centrality to Lifestyle, measures how relevant an activity is to a person’s life, and the third, Symbolic Value, reflects the self-expression level an activity represents to an individual. Each facet was measured with a 3item scale that was determined to be valid assessments of sports involvement. Furthermore, the authors cited research that suggested the multidimensional scale would uncover more information about involvement than a unidimensional scale (Havitz & Dimanche, 1997; Kapferer & Laurent, 1993; Laurent & Kapferer, 1985). The 2011 study on sports involvement was used to assess the factors that correlate with changes in the amount of leisure activities by marathon runners who belonged to running clubs. The current project will assess sports fans’ level of sports media involvement in order to determine if significant correlations exist with changes in levels of sports media consumption per day. Motivational Antecedents of Involvement As discussed in the previous chapter, the degree of involvement perceived by an individual is determined by the level of personal importance and relevance of an object or activity (Johnson & Eagly, 1989). Alongside this, there is an assumption to the involvement construct that a consumer will have specific motivations that help determine the level of involvement with the object or activity (Antil, 1984). A meta-analysis of involvement research explicitly stated that the concept of involvement can be defined in terms of the motivations that precede it (Costley, 1988). Applicable to the current study, literature on leisure involvement has made particular contributions to the exploration of the motivations behind involved leisure behavior (Bloch & Bruce, 1984). The current study adapts sports involvement (Beaton et al., 2011), which based the bulk of its measurement items on studies of leisure involvement (Kyle & Mowen, 2005; McIntyre, 1989). As media consumption in the form of television watching is 27 considered a leisure activity by the United States Bureau of Labor Statistics (Bureau of Labor Statistics, 2014, 2015), it can be said that modifying scales of leisure involvement, or in this study’s case, sports involvement (Beaton et al., 2011), to measure sports media involvement is a valid measure of the construct. The earliest examination of human motivations was conducted by Abraham H. Maslow in order to explain the basic needs of all human beings (Maslow, 1943), and has been adapted in the years following (Maslow, Frager, Fadiman, McReynolds, & Cox, 1970). More recently, it was research into leisure involvement that initiated the assessment of motivations to pursue leisure activities. One study explained that motivations are the fundamental reasons for behavior and that a combination of intrinsic motivations facilitates attempts to achieve optimum levels of arousal from pleasure-seeking activities (Iso-Ahola, 1989). Consumption of sports media fits this profile since it is also a leisure activity. Individuals’ frequency of participation, or time spent consuming sports media, and level of involvement are affected by a combination of intrinsic rewards garnered from psychological motivations (Carroll & Alexandris, 1997; Graef, Csikszentmihalyi, & Gianinno, 1983). This is why it is important to assess the discriminant driving motivations of the 21st Century sports media consumer; to be able to provide a more complete understanding of which types of reward-seeking desires correspond with changes in sports media consumption. The earliest documented scale measuring motivations was conducted in order to help determine why people sought to perform leisure activities (Beard & Ragheb, 1983). Items were developed and tested to understand leisure behavior and the major dimensions of leisure motivations. The results of the study found four reliable subscales consisting of 12 items each. The authors labeled the four factors of leisure motivations as intellectual, social, mastery- 28 competence, and stimulus avoidance. These subscales served as the foundation for future research, as will be detailed in the remainder of this chapter. Of particular importance is the recognition that there exist types of motivations that relate to both the seeking of individual rewards and the escaping from external environments (Carroll & Alexandris, 1997). Ten years after the 1983 leisure motivation scale by Beard & Ragheb was developed, the Sport Commitment Model (SCM) was developed to assess the motivations essential to continued sport involvement in youth sports activities (Scanlan, Simons, Carpenter, Schmidt, & Keeler, 1993a). The measures were established in order to pinpoint the desires behind continued sports participation. This commitment to participate was defined as a state of attachment reflecting the motivational forces for continued involvement. Five determinants were hypothesized to affect the level of sport commitment. These concepts were sports enjoyment, involvement alternatives, personal investments, social constraints, and involvement opportunities (Scanlan, Simons, Carpenter, Schmidt, & Keeler, 1993b). Three of the five core concepts – sports enjoyment, social constraints, and involvement opportunities, were found to significantly affect sports commitment in a positive direction. The other two variables – involvement alternatives and personal investments, were removed from the model due to measurement issues of internal consistency. The SCM has been used in more recent studies to assess the participation frequency and purchase intentions of adult tennis players (Casper, Gray, & Babkes Stellino, 2007). That study employed all five of the original SCM core concepts, determining that enjoyment and personal investments were the two main motivations influencing continued participation and purchase intention for adult recreational tennis players. Another scale of motivation was developed in order to explore intrinsic and extrinsic motivations of athletes. The SMS or Sport Motivation Scale was adapted from a French study 29 seeking to understand the driving forces motivating collegiate-aged soccer players in Ottawa, Canada (Pelletier et al., 1995). Seven reliable and valid subscales were found, but these finding are not applicable to the current study because the sample population was on athletes and not sports spectators or consumers. A more recent study used the SMS to examine participation motivations of amateur sports participants in community centers in London, England (Vlachopoulos, Karageorghis, & Terry, 2000). Another motivation study from this time introduced the Sports Fan Motivation Scale or SFMS (Wann, 1995). The SFMS was created to measure the motivations of sports spectators and television viewers, referred to as sports fans. The factors were believed to be the responsible motivations behind sports fandom. The study is important because the various categories of motivations in the SFMS influenced or were directly used as the specific types of motivations studied in future literature investigating motivational drivers of involvement and sports consumption (Armstrong, 2002a, 2002b; Funk, Mahony, Nakazawa, & Hirakawa, 2001; Funk et al., 2002; Funk et al., 2004; Hall, O’Mahony, & Vieceli, 2010; Izzo et al., 2011; M. J. Robinson & Trail, 2005; Trail et al., 2003; Trail & James, 2001; Wann, Grieve, Zapalac, & Pease, 2008). Furthermore, each of the factors are related in some way to the different motivational variables tested previously by this author in an attempt to predict changes in sports media consumption per day (Krier, 2016). They will also be examined in that fashion in the current study. The SFMS consisted of eight subscales of motivation to watch sports: eustress (emotional stimulation), self-esteem, escape from problems, entertainment, economic (betting on sports), aesthetic (artistic value), group affiliation (bonding with friends), and family needs (Wann, 1995). It was determined that each two or three-item subscale was reliable and that each subscale showed significant correlations with one another. The highest scoring motivation was discovered 30 to be entertainment and the lowest scoring motivation was economic or betting on sports. Significant gender differences were found in seven of the eight motivations, with men scoring significantly higher on six of the eight (females scored higher on the family motivation). The generalizability of these findings is suspect for current sports media consumers since it is over twenty years old and current statistics show that females have increased in sports fan viewership percentages over the years to nearly equal numbers as males (Sporting News Media, 2014). In 2001, two separate but similar sports motivation scales were developed to test the psychological drivers of sports spectators. The first, published in March 2001, was called the Motivation Scale for Sports Consumption (MSSC). It was created in order to assess the motives behind sports fans who invested time, money, and emotional consequence to watch and follow sports (Trail & James, 2001). The specific motivations were based on Wann’s 1995 SFMS as well as sports sociology literature (Sloan, 1989) and a previous study on college football fan attendance (Kahle, Kambara, & Rose, 1996). It was determined by Trail & James that many of the factors from the SFMS lacked various psychometric properties and were not valid and reliable measures of sports spectator motivations. Testing season ticket holders of an NBA basketball team, the MSSC discovered nine discrete different motivations to attend: vicarious achievement, acquisition of knowledge, aesthetics, drama, escape, family, physical attraction to players, physical skills of players, and social interaction. The authors pointed out that the contribution of their study was the discovery of the two new factors of acquisition of knowledge and physical attraction of the players. Many of these motivational factors are tested in the current study. In June 2001, another study was published dealing with the motivations of sports spectators; this time testing individuals who attended the 1999 FIFA Women’s World Cup (Funk 31 et al., 2001). The Sports Interest Inventory (SII) aimed to identity of the underlying motivations to watch sports (Funk et al., 2002). Like the MSSC, the study based its factors on Wann’s 1995 SFMS as well as Kahle et al., 1996. The results produced 10 psychometrically sound factors, some of which are specific to testing motives for watching women’s soccer games. The 10 factors are as follows: interest in soccer, vicarious achievement, excitement, team identification, support of women’s opportunities in sports, aesthetics, socialization, national pride, drama, and interest in players. Of note here is that team identification is considered a motivation for consuming sports. While this might be the case, the SII is the only scale of motivations that incorporates team identity. A recent study by the current author discovered team identity to be a separate construct that not only significantly predicts increases in other motivations to consume sports media, it was also found that higher identity itself predicts increased consumption with the motivations serving as mediator variables (Krier, 2016). The SII factors from 2001 went on to be used in future studies to investigate gender differences in sports spectators (M. J. Robinson & Trail, 2005) as well as online sports consumers (Seo & Green, 2008). One study modified both the MSSC and the SII to explore the spectator motivations of Romanian soccer fans (Izzo et al., 2011). To conclude their 2001 paper, the authors of the SII study suggested that further investigation into the motives of sports fans was necessary to better understand the underlying motives to consume. That is exactly what was done when the SII was administered to season ticket holding sports fans of a professional team in order to determine which motives served as origins to four facets of involvement (Funk et al., 2004). The facets of involvement were taken from the Involvement Profile or IP Scale of leisure involvement (Havitz & Dimanche, 1997, 1999; Havitz & Howard, 1995; Iwasaki & Havitz, 2004; Laurent & Kapferer, 1985). The four 32 facets consisted of attraction, self-expression, centrality to lifestyle, and risk. The 2004 Funk et al. study employed confirmatory factor analysis to test the significance of the relationship between each discrete motivational factor and its respective involvement construct. The 18 separate involvement antecedents tested in the study were as follows: role models, team interest, support for women’s opportunities, entertainment value, excitement, wholesome environment, drama, style of play, basketball knowledge, customer service, family bonding, vicarious achievement, interest in basketball, bonding with friends, socialization, community pride, escape from problems, and interest in players. While each involvement antecedent fit the data in the confirmatory factor loadings, there was no structural model employed to test these motivational relationships to involvement with any behavioral outcomes. The authors noted in their limitations section that future research should look to test the model with specific sports consumption outcomes. That is what this current study is seeking to accomplish. The 18 origins to involvement were pared down to 12 factors for this project predicting changes in total sports media consumption per day. Each of the factors were modified from valid and reliable scales to apply to sports media consumption. They are tested in conjunction with a scale measuring team identity and one of sports media involvement to predict changes in daily sports media consumption. Table 1 below lists the reliability of the constructs being measured in this research project. 33 Construct Table 1: Reliability of Constructs Variable Name # of Items Chronbach’s α Source Unidimensional Team Identity Team Identity Level 4 0.877 (Wann & Branscombe, 1993) 3 0.943 (Heere et al., 2011) via (Gurin & Townsend, 1986) Public Evaluation 3 0.869 (Heere et al., 2011) via (Luhtanen & Crocker, 1992) Private Evaluation 3 0.914 (Heere et al., 2011) via (Luhtanen & Crocker, 1992) Interdependence of Self w the Group 3 0.865 (Heere et al., 2011) via (F. A. Mael & Tetrick, 1992) Behavioral Involvement 3 0.895 (Heere et al., 2011) via (Phinney, 1992) Cognitive Awareness Role Model (ROL) 3 0.912 (Heere et al., 2011) 3 0.93 (Funk, Ridinger, & Moorman, 2004) Team Interest (TEM) Excitement (EXC) Drama (DRA) Basketball Knowledge (KNW) Bonding with Family (FAM) Vicarious Achievement (VIC) 3 0.88 3 0.90 3 3 0.79 0.85 3 0.92 3 0.92 Multidimensional Sense of Interdependence Team Identity Motivations to Consume Sports Media 34 Table 1 (cont’d) Sports Involvement Bonding with Friends (BON) Socialization (SOC) Community Pride (COM) Escape (ESC) Interest in Players (PLA) Hedonic Value 3 0.83 3 0.85 3 0.85 3 3 0.93 0.79 3 0.86 Symbolic Value Centrality to Lifestyle 3 3 0.86 0.82 35 (Beaton et al., 2011) Linking Identity Level to Motivations to Consume, Involvement, and Sports Media Consumption “The degree to which one is committed to a social identity determines the power of that identity to influence behavior” (Solomon, M. R., 1983, p.321). The quote speaks on the level of one’s identity in the context of products and social behavior, and provides a direct link between identity level and behavior (Solomon, 1983). As discussed in the introduction to this paper, sports and sports media are multi-billion dollar industries both domestic and worldwide (Statista, 2016e; 2016b). The volume of time consuming sports and the amount of money spent and earned from media productions of sports are similar to other types of products in that they can be consumed by individuals at differing levels (Bennett, Sagas, & Dees, 2006; L’Etang, 2006; McCarthy, 2012). Research has noted that products and services can be labelled as either sportsrelated or non-sports-related (Fullerton & Merz, 2008). As research has stated that higher levels of team identification result in a willingness to engage and participate in consumption behaviors that support the group (Gwinner & Swanson, 2003), it is logical to assume the level of team identity felt by an individual will have a correlation with the behavior of consuming sports media. The concept of involvement is even more closely tied to products and consumer behavior (Mitchell, 1979; Zaichkowsky, 1985, 1986), as it reflects the self-relevance of an activity or object to an individual (Johnson & Eagly, 1989; Rothschild, 1984). Involvement has been studied in various types of research regarding sports fan behaviors such as purchasing merchandise and attending live sporting events (Beaton & Funk, 2008; Bloch & Bruce, 1984; Casper & Stellino, 2008; Funk et al., 2004; Hill & Green, 2000; Walker & Kent, 2009). The 36 current study will test the correlation between one’s level of sports media involvement and daily consumption of sports media. The previous section detailed the concept of motivations of behavior as antecedents to involvement, especially as sports consumption motivations relate to involvement (Funk et al., 2004). Motivations are said to comprise the specific details that help define what one can consider involvement (Antil, 1984; Costley, 1988). Therefore, the link between behavioral motivations and sports media involvement is a logical connection. Attitude has been defined as a tendency that is psychologically expressed by individuals regarding the positive or negative evaluation of a particular entity (Eagly & Chaiken, 2007). As discussed in the section of this paper addressing involvement, existing research has explained that social identity is an antecedent to attitude importance (Boninger et al., 1995). According to that study, attitude can be operationalized in such a way that “emphasizes a subjective sense of psychological significance…best measured by means of people’s self-reports” (Boninger et al., 1995, p. 62). Similarly, SIT explains one’s sense of identity as being a consequence of a knowledge of group membership with attached emotional significance (Tajfel, 1978). As involvement has been shown to affect the magnitude to which someone will change his or her attitude (Barki & Hartwick, 1994; Kokkinaki & Lunt, 1997; Sherif et al., 1965), a link between social identity and involvement is established. Additional links between social identity and involvement were discussed in the earlier section. Increased level of involvement was shown to strongly correlate with increased levels of team identity (Fisher & Wakefield, 1998), and the overall sports marketplace has been investigated by exploring the relationship between involvement and social identity (Underwood et al., 2001). Due to the bulk of theoretical associations presented in this chapter, the logic 37 behind the predictive structural equation model in the current study is established. As such, it will explore the correlations between team identity level, motivations to consume, sports media involvement, and the amount of time spent consuming sports media. The Dependent Variable of Sports Media Consumption in Total Hours per Day Media consumption is defined as the total amount of informational media taken in via reading, watching, or listening to a source by an individual or a group of people. Measuring this construct via a time variable has been said to be an effective and appropriate manner in which to analyze consumption habits of individuals (J. P. Robinson, Kestnbaum, Neustadtl, & Alvarez, 2000). Specifically listing three useful features of measuring time spent consuming, researchers of that study noted about time that, “First, its measurement units are universally recognized and used as benchmarks for daily living…Second, time has become recognized as an important societal resource that needs to be managed…Third, it is a resource that is equally distributed to all members of society” (J. P. Robinson et al., 2000, p. 491). The question of what it means to consume a fixed amount of time on one medium as opposed to another is a difficult one to answer, as there has been little research conducted directly comparing amounts of information consumed across media types. Studies on media effects are wide-ranging, though, from Cultivation Theory and the effects of violent media (Gerbner, 1998), the work of Zillmann & Bryant on the hedonic effects of media consumption (Zillmann & Bryant, 1988), to studies on whether media type affects perceptions of female body image (Bell & Dittmar, 2011). While these studies are all appropriate for examining the outcomes of media consumption, the current study is focused on the antecedent drivers of sports media consumption, specifically regarding social identity and involvement. 38 Other studies have looked at the existence of new technology like decentralized online media as it relates to existing technology and old, centralized media like broadcast television and radio or print. The question is whether the use of new media displaces or supplements what already exists, and research shows that displacement is not simply occurring. Instead a shift of perception of the purpose of media has transpired (Kayany & Yelsma, 2000) and fear of new media cannibalizing old media is unsupported (Ahlers, 2006). Some research has answered the question by illustrating that people use newer media such as Internet news sources to supplement their other media consumption, not replace it (Althaus & Tewksbury, 2000; Chan & Leung, 2005). Other research points out that displacement theory is not as easily applicable to new media because it is not a zero-sum game where the use of one medium directly replaces that of another (Dutta-Bergman, 2004). That 2004 paper points out the difficulty of assessing media types as being homogenous due to the fact that two people can consume the same type of media for the same amount of time, but for different purposes. Additionally, a recent study detailed that there are many different findings in existing literature regarding whether new media is in fact displacing old media in any consistent fashion (Chyi & Lee, 2013). This is because different individuals do not have the same experiences using any certain media type, which makes it difficult to assess what it means to consume different types of media. While there does not exist a conversion table to equate such things as reading texts of sports media as opposed to watching a video or live broadcast of sports media, some studies have found that attention spans can be similar comparing different types of media. Such research measured visual attention paid to news stories broadcast on television versus written on a computer screen and determined that extended gazes happened on the computer 54.9% of the 39 time as opposed to a similar number of 48.9% of the time for televised media (Brasel & Gips, 2011). Simultaneous media use has been growing at a steady pace for Americans in the 21st Century due to the availability of new and more easily accessible technology (Brasel & Gips, 2011; Ophir, Nass, & Wagner, 2009). As was stated in the previous two paragraphs, this is because using new media does not necessarily displace other media being consumed, but rather supplements it (Althaus & Tewksbury, 2000; Chan & Leung, 2005), creating what has become known as media multitasking. Media Multitasking is defined as engaging in simultaneous use of more than one media activity at one time, or multiple exposures to various media at one single point in time (Pilotta, Schultz, Drenik, & Rist, 2004; Voorveld & van der Goot, 2013). Studies measuring the simultaneous use of more than one type of media have defined two constructs to attempt to show a difference in time spent consuming media: media exposure and media use (Foehr, 2006; Roberts & Foehr, 2008). Total media exposure is defined as the sum of all time spent consuming media while total media use is the actual amount of time in one day spent using media, accounting for multitasking (Rideout, Foehr, & Roberts, 2010). Using diaries of sampled individuals to help determine a time value comparison of the constructs, two studies listed media use times that averaged 78% less than the total time of media exposure (Rideout et al., 2010; Roberts & Foehr, 2008). These studies noted, however, that without concurrent measurement of multitasking during simultaneous media exposure, an accurate value of media use cannot be validly attained. Other studies have taken a different approach to measuring simultaneous media usage. One study in particular measured time spent consuming more than one type of media simultaneously as a sum of all times for each specific medium (Pea et al., 2012). This is the same 40 method the current study will employ to measure time spent consuming sports media per day. While differences have been found for various age groups regarding time spent consuming media simultaneously, with youngsters aged 13-16 predominantly using the Internet and social media simultaneously while adults aged 50-65 used television and newspapers simultaneously most often (Voorveld & van der Goot, 2013), that study also noted that the media landscape is shifting so quickly due to new technology that longitudinal studies are necessary in order to achieve a more accurate view of modern media consumption. There is little doubt that measuring time spent consuming media is of great worth in the present field of media research. Regarding the current study, it has been stated that people with a heightened sense of involvement in sports, i.e. interest in activities such as watching or following sports media, could lead to increased time consuming sports media for those individuals (Funk et al., 2004; Tsiotsou, 2013). This is due to the increased attention devoted to a message source in the presence of high involvement, which has been said to impact behavior (Greenwald & Leavitt, 1984) and result in behavioral effects such as higher frequency of usage (Mittal & Lee, 1989). Additionally, team identity level was found to have a direct influence on levels of sports media consumption (Heere et al., 2011). The current study will measure the effects of team identity level and involvement on the outcome variable of time spent consuming sports media per day. This dependent variable is a sum of self-reported hours spent watching or following sports on television, the radio, the Internet (on computers and Smartphones), reading newspaper sports sections or sports magazines, and social media platforms, measured in 1-hour intervals for each type of medium. A large amount of existing literature independently examining social identity, involvement, and motivations have used attendance at events as the dependent variable to assess a variety of effects from the different constructs. Regarding identity, it was shown that a 41 consequence of higher levels of identification with an organization results in higher attendance at functions associated with that group (Ashforth & Mael, 1989). Sports team identification specifically has been shown to affect attendance decisions (Dimmock, Grove, & Eklund, 2005), even allowing for the prediction of behaviors such as number of games attended (Arnett & Laverie, 2000; Fisher & Wakefield, 1998; Laverie & Arnett, 2000; Madrigal, 2000; Wann et al., 1994). Overall, the conclusion is that more highly identified sports fans are more involved with the team, thus resulting in an increased likelihood of game attendance (Underwood et al., 2001; Wann & Branscombe, 1990, 1993). Additionally, frequency of attendance at sporting events was shown to be positively correlated with increased levels of involvement (Hill & Green, 2000; Vlachopoulos et al., 2000). Motivational antecedents to involvement were also found to strongly influence sporting event attendance (Hall et al., 2010; M. J. Robinson & Trail, 2005), illustrating that increased involvement correlates with increased attendance. Higher levels of motivations to attend games of a particular team were also shown to be correlated with an increase in consumption of Internet content for those teams’ official websites (Seo & Green, 2008). This last point illustrates that it is logical to predict that higher levels of involvement will correlate with more time spent consuming sports media. The specific measurement of time spent consuming media in minutes and hours per day is prevalent in existing news media research. A study on the patterns of traditional and online news media consumption measured self-reported hours per day spent reading the newspaper versus online news media sources (Althaus & Tewksbury, 2000). Similar studies on news media usage habits have placed the measurement of hours consuming online news media into ordinal categories for assessment (Dutta-Bergman, 2004; Strömbäck, Djerf-Pierre, & Shehata, 2012). A 42 study on mobile news adoption measured time spent in minutes using mobile news per occasion on a mobile device (Chan-Olmsted, Rim, & Zerba, 2013). One particular study used existing Pew Center survey research that had queried households on their usage habits on a wide variety of news media, measuring media consumption in minutes per day (J. P. Robinson et al., 2000). Lastly, three studies this decade have all measured time spent consuming news media in the form of either self-reported minutes per day (Yuan, 2011), half-hour intervals in order to operationalize the construct of audience availability (Taneja, Webster, Malthouse, & Ksiazek, 2012), and one-hour intervals to explain patterns of news consumption across various types of media (Ksiazek, Malthouse, & Webster, 2010). As mentioned in the introduction of this paper, sports media consumption is a daily occurrence for 168 million Americans, or roughly 70% of the United States population (Sporting News Media, 2014). According to that 2014 study, 96% of sports fans consume their sports on television, 68% follow sports online on a daily basis, and 35% consume sports via social networking platforms, with 42% of fans consuming content via a mobile device such as a Smartphone or Tablet. The objective of the current research project is to determine the extent to which the variables of team identity, motivations to consume, and sports media involvement correlate and predict changes in the amount of sports media consumed. Therefore, the dependent variable in this research project is Total Sports Media Consumed per Day measured in cumulative hours consumed. Total Consumption is a sum of each individual’s self-reported hours spent consuming sports media per day on television, the radio, the Internet (on computers and Smartphones), reading newspaper sports sections or sports magazines, and on social media platforms. Besides watching television in a traditional sense, this total will include things like streaming live or 43 recorded sporting events online, or watching sports clips and videos on the Internet via YouTube. Radio includes listening to podcasts, and social media use includes anything dealing with sports over platforms such as Facebook, Twitter, Snapchat, and Google+, all of which are listed as the top social networks via which fans consume sports media (Sporting News Media, 2014). It should be noted that the measured amounts of time spent consuming sports media reflects the psychological and sociological importance of sports to individual sports media consumers in the United States. These individuals consuming sports media are behaviorally illustrating the importance of their levels of team identity and the amount of personal relevance felt with sports media as reflected by their levels of involvement. As discussed earlier, the social and psychological aspects of sports participation and spectatorship result in such characteristics of fan identification which include enhanced levels of self-esteem, reduced feelings of depression, and a lowered sense of alienation from the rest of society (Branscombe & Wann, 1991). Sports media consumption is a mediated form of sports spectatorship. The structural model tested in this study attempts to predict the amount of total variance explained of the dependent variable of total sports media consumption per day by the independent variables of team identity level, 12 discrete antecedent motivations of involvement, and level of sports media involvement. Research Model, Hypotheses, and Research Questions The theoretical framework for the current research model combines Social Identity Theory in the form of sports team identity level (Heere & James, 2007b; Heere et al., 2011; Wann & Branscombe, 1993) with various motivational antecedents of involvement and the consumption of sports media, adapted from a study on motivations to attend a live sporting event (Funk et al., 2004), and the construct of sports media involvement based on sports involvement 44 (Beaton et al., 2011). The objective is to determine if factors developed for predicting sporting event attendance and sports involvement for marathon runners apply to sports media consumption. 96% of sports fans consume sports by watching television, 68% use the Internet, 42% use mobile devices, and 41% listen to some type of sports radio (Statista, 2014b). For these reasons, the dependent variable is time spent consuming sports media, measured in hours per day. In this study, total consumption is a sum of self-reported hours spent consuming sports media per day on television, newspapers & magazines, the radio, the Internet (on computers and Smartphones), and social media platforms like Twitter, Snapchat, and Facebook. The predictive structural equation models of the current study seen below in Figures 1 and 2 will explore the correlations between team identity level, antecedent motivations, sports media involvement, and the amount of time spent consuming sports media. Theoretical links exist regarding each of the constructs, and guide the hypotheses and research questions of this study. Of note, this project will compare two similar models differing in the constructs of multidimensional team identity – six 3-item scales – (Heere & James, 2007b) versus unidimensional team identity – a 4-item scale – (Wann & Branscombe, 1993) and multidimensional sports media involvement – three 3-item scales – (Beaton et al., 2011) versus unidimensional sports media involvement – a 5-item scale – (developed by this study) in regard to variance explained of the dependent variable of total sports media consumption per day. Additionally, the models will test for mediation of the effect of team identity on total consumption by the variables of antecedent motivations and sports media involvement, with analysis recommendations directed by extant literature (Baron & Kenny, 1986; MacKinnon, Fairchild, & Fritz, 2006). Overall, this leads to the subsequent hypotheses: 45 As the degree of one’s social identity has been said to determine the power of such identity to influence related behaviors (Solomon, 1983), the following hypothesis is posited: H1: Higher levels of Team Identity will result in higher levels of Total Sports Media Consumption per day. Existing research notes that social identity with a particular group is an antecedent to attitude importance (Boninger et al., 1995). Increased involvement has been shown to affect the extent to which someone is willing to change his or her attitude towards a product or service (Barki & Hartwick, 1994; Kokkinaki & Lunt, 1997; Sherif et al., 1965). Therefore, one can conclude that level of identity determines the level of one’s attitude acceptance or rejection, or personal relevance, establishing a link between identity and involvement. Additional research has noted that a positive attitude correlates with increased consumption intention regarding increased social identity (Madrigal, 2001). Therefore, the following hypothesis is posited: H2: Higher levels of Team Identity will result in higher levels of Sports Media Involvement. As the concept of motivations to consume are established as antecedents to involvement (Antil, 1984; Costley, 1988), especially as sports consumption motivations relate to involvement (Funk et al., 2004), it is logical to assume the rationale for H2 will apply to a statistical association between team identity level and motivations to consume sports media. Furthermore, it was discovered in a prior study by this author that increases in team identity do correlate with increases in significant motivational drivers of total sports media consumption (Krier, 2016). Therefore, the following hypothesis is posited: H3: Higher levels of Team Identity will result in higher levels of significant predictor Antecedent Motivations. 46 Being that the concept of motivations to attend sporting events is established as antecedents to involvement (Funk et al., 2004), there is a clear link between motivations to consume and sports media involvement. Therefore, the following hypothesis is posited: H4: Higher levels of significant predictor Antecedent Motivations will result in higher levels of Sports Media Involvement. The prior study by this author on the constructs of motivations to consume sports media and total sports media consumption per day revealed that increases in three significant motivational factors predicted increases in time spent consuming (Krier, 2016). Therefore, the following hypothesis is posited: H5: Higher levels of Antecedent Motivations will result in higher levels of Total Sports Media Consumption per day. Research has shown the concept of involvement to be closely related to consumer behavior (Mitchell, 1979; Zaichkowsky, 1985, 1986). Involvement literature includes studies on sports fan behaviors linking increases in involvement with increased purchasing of team-related merchandise and attending live sporting events (Beaton & Funk, 2008; Bloch & Bruce, 1984; Casper & Stellino, 2008; Funk et al., 2004; Hill & Green, 2000; Walker & Kent, 2009). Since consuming sports media is another possible behavioral outcome of increased involvement, the following hypothesis is posited: H6: Higher levels of Sports Media Involvement will result in higher levels of Total Sports Media Consumption per day. Research has shown that involvement can act as a mediating variable between attitudes and actions regarding consumer behavior (Mitchell, 1979; Ray, 1979; Zaichkowsky, 1985). Therefore, it is likely that the intermediary variables in the model will contribute some type of 47 mediating effect, full or partial, on the direct influences between constructs (Hall et al., 2010). Following the analytical recommendations of prior research (Baron & Kenny, 1986; MacKinnon et al., 2006), the following hypotheses on mediation are posited: H7: Sports Media Involvement mediates the relationship between Team Identity level and Total Sports Media Consumption per Day. H8: Antecedent Motivations mediate the relationship between Team Identity and Sports Media Involvement. H9: Antecedent Motivations mediate the relationship between Team Identity and Total Sports Media Consumption per Day. Next, as noted earlier in this section, in order to test the effectiveness of explaining the variance of total sports media consumption explained, two different structural models were tested that are similar except for their measures of team identity and sports media involvement (multidimensional vs. unidimensional). The specific items are detailed in Table 2 below. Model #1 employed multidimensional team identity – consisting of six 3-item scales – (Heere & James, 2007b) and multidimensional sports media involvement – three 3-item scales – (Beaton et al., 2011) while Model #2 tested unidimensional team identity – a 4-item scale – (Wann & Branscombe, 1993) and unidimensional sports media involvement – a 5-item scale developed from the data of this project. Therefore, the following research question is posited: RQ1: Does Model #1 with Multidimensional Team Identity (Heere & James, 2007b) and Multidimensional Sports Media Involvement (Beaton et al., 2011) explain more variance of Total Sports Media Consumption per day than Model #2 using Unidimensional Team Identity (Wann & Branscombe, 1993) and Unidimensional Sports media involvement? 48 As the six first-order facets of Multidimensional Team Identity – public evaluation, private evaluation, interconnection of self with the team, sense of interdependence with the team, behavioral involvement, and cognitive awareness – will uncover the degree to which each factor contributes to measuring team identity level (Ashmore et al., 2004; Heere & James, 2007b), the following research question is posited: RQ2: Which first-order facets of Multidimensional Team Identity contribute most to the identity construct? As the three first-order facets of Multidimensional Sports Media Involvement – hedonic value, centrality to life, and symbolic value – will uncover the degree to which each factor contributes to measuring level of involvement (Beaton et al., 2011), the following research question is posited: RQ3: Which first-order facets of Multidimensional Sports Media Involvement contribute most to the involvement construct? Finally, as discussed in the introduction, female sports media consumers constitute a large proportion of sports fans, with 58% of women identifying as sports fans in 2002 (Bush, Bush, Clark, & Bush, 2005). By 2012, 63% of women classified themselves as sports fans, and 46% of adult females regularly viewed NFL football games on television each week (Adweek, 2014). Therefore, an investigation into any gender differences in the structural model is of great importance to determine if there are any cultural differences in sports media consumption based on gender. Therefore, the following research question is posited: RQ4: Are there significant gender differences in the motivational drivers of sports media involvement and sports media consumption? 49 Figures 1 and 2 illustrate the structural models of the current study with hypothesis numbers listed next to the corresponding arrows: Figure 1: Proposed Model #1 of Multidimensional Constructs Proposed Model #1 of Multidimensional Team Identity based on Heere & James (2007); Heere et al. (2011), Motivations to Consume based on Funk et al. (2004), and Sports Media Involvement based on Beaton et al. (2011) as predictors of Total Sports Media Consumption per Day (in hours). Figure 2: Proposed Model #2 of Unidimensional Constructs Proposed Model #2 of Unidimensional Team Identity based on Wann & Branscombe (1993), Motivations to Consume based on Funk et al. (2004), and Unidimensional Sports Media Involvement based on Beaton et al. (2011) as predictors of Total Sports Media Consumption per Day (in hours). 50 In order to measure the relationships between the constructs in the above models, a survey questionnaire of 72 items was administered to United States adults over 18 who regularly watch or follow sports media. Table 2 provides the operationalization of constructs. Table 2: Operationalization of Constructs NOTE: “Sports media” includes live or recorded sporting events, sports news or highlights, sports talk shows or podcasts, social media for sports (via Twitter & Facebook), and/or sports programming such as sports documentaries. Construct Unidimensional Team Identity (TI_U) Item 4 TI_U items TI_U1 TI_U2 TI_U3 TI_U4 Multidimensional 18 TI_M Items Team Identity (TI_M) Public Evaluation TI_M_PbE1 TI_M_PbE2 TI_M_PbE3 Question Feelings of Unidimensional Team Identity (1=Strongly Disagree; 7= Strongly Agree) How important to you is it that your favorite sports team wins? How strongly do you see yourself as a fan of your favorite sports team? How strongly do your friends see you as a fan of your favorite sports team? How important is being a fan of your favorite sports team to you? Feelings of Multidimensional Team Identity (1=Strongly Disagree; 7= Strongly Agree) Overall, my favorite team is viewed positively by other people. In general, other people respect my favorite team. People hold favorable opinions about my favorite team. 51 Source (Wann & Branscombe, 1993) (Heere et al., 2011) (Heere et al., 2011) via (Luthanen & Crocker, 1992) Table 2 (cont’d) Private Evaluation TI_M_PvE1 TI_M_PvE2 TI_M_PvE3 Sense of Interdependence with the Group TI_M_IntD1 TI_M_IntD2 TI_M_IntD3 Interconnection with the Group TI_M_IntC1 TI_M_IntC2 TI_M_IntC3 Behavioral Involvement TI_M_B1 TI_M_B2 TI_M_B3 Cognitive Awareness TI_M_CA1 I feel good about cheering for my favorite team. In general, I’m glad to be a fan of my favorite team. I am proud to think of myself as a fan of my favorite team. The performance of my favorite team influences what happens in my life. Changes that affect my favorite team have an impact on my life. What happens to my favorite team has an impact on my life. When someone criticizes my favorite team, it feels like a personal insult. Being associated with my favorite team is an important part of my selfimage. When someone says something positive about my favorite team, it feels like a personal compliment to me. I participate in activities supporting my favorite team. I am actively involved with any activities related to my favorite team. I participate in activities with other fans of my favorite team. I am aware of the traditions of my favorite team. 52 (Heere et al., 2011) via (Luthanen & Crocker, 1992) (Heere et al., 2011) via (Gurin & Townsend, 1986) (Heere et al., 2011) via (Mael & Tetrick, 1992) (Heere et al., 2011) via (Phinney, 1992) (Heere et al., 2011) Table 2 (cont’d) TI_M_CA2 TI_M_CA3 Motivations to Consume Sports Media 36 Motivation items PLA1 (Interest in Players) PLA2 (Interest in Players) PLA3 (Interest in Players) BON1 (Bonding with Friends) BON2 (Bonding with Friends) BON3 (Bonding with Friends) DRA1 (Drama - Sports Media Updates) DRA2 (Drama - Sports Media Updates) DRA3 (Drama - Sports Media Updates) I am aware of the specific details regarding my favorite team (e.g.: history, win/loss record, current and past players, coaches, etc.). I am familiar with the past successes and failures of my favorite team. Feelings regarding (Funk, Ridinger, & Motivations to Consume Moorman, 2004) Sports Media (1=Strongly Disagree; 7= Strongly Agree) I cheer for one or more athletes that I consider my favorite. I will follow my favorite player no matter which team he or she plays on. I like my favorite player more than the team he or she plays for. I feel that sports media help me bond with my friends. Sports media allow for a shared experience for me with my friends. Sports are more enjoyable when my friends and I are watching together. I enjoy getting live sports updates through sports media. I like checking sports media for updates about live games. I like seeing sports media alerts about breaking news. 53 Table 2 (cont’d) COM1 (Community Connectedness) Sports media make me feel connected to the sports fan community. The various types of sports media help me to feel part of the sports fan community. Sports media are COM3 (Community important in my life Connectedness) because they connect me to the sports fan community. I feel that collegiate and ROL1 (Role Models) professional athletes are good role models for me and others. Successes by other ROL2 (Role Models) athletes inspires me to become more successful on my own. Athletes are people that ROL3 (Role Models) inspire me to be a better person. Watching and following EXC1 (Exciting Coverage sports media is exciting of my Team) to me. Sports media help to EXC2 (Exciting Coverage make sports more of my Team) exciting. Sports media provide a EXC3 (Exciting Coverage constant source of of my Team) excitement in my life. COM2 (Community Connectedness) KNO1 (Knowledge of Sports) KNO2 (Knowledge of Sports) Sports media help to expand my knowledge about sports in general. I like when sports media provide me with statistical information about players, teams, and game stats. 54 Table 2 (cont’d) KNO3 (Knowledge of Sports) VIC1 (Vicarious Achievement) VIC2 (Vicarious Achievement) VIC3 (Vicarious Achievement) SOC1 (Social Media Interaction) SOC2 (Social Media Interaction) SOC3 (Social Media Interaction) ESC1 (Escape from Problems) ESC2 (Escape from Problems) ESC3 (Escape from Problems) The information provided by sports media make sports more enjoyable to me. Sports media that focuses on winners and champions makes me feel like a winner too. Sports media makes me feel a sense of personal accomplishment when I experience sports media coverage of winners and champions. Sports media coverage of success in sports is the main reason why I watch and follow sports. I like when sports media allow me to interact with other fans via social media. I like communicating with other sports fans in an open public forum via social media. I like when sports media allow me to voice my opinion about players and teams. Sports media distract me from the bad or annoying things in my everyday life. Sports media are part of my daily or weekly routine because they allow me to escape my problems. I can get away from the problems in my life by watching sports or following sports media. 55 Table 2 (cont’d) FAM1 (Family Atmosphere) FAM2 (Family Atmosphere) FAM3 (Family Atmosphere) TEM1 (Team Interest – Behind the Scenes) TEM2 (Team Interest – Behind the Scenes) TEM3 (Team Interest – Behind the Scenes) Multidimensional 9 SP_INV_M Items Sports Media Involvement (SP_INV_M) Hedonic Value SP_INV_M_HV1 SP_INV_M_HV2 SP_INV_M_HV3 Centrality to Lifestyle SP_INV_M_C1 Sports media creates a family-type atmosphere when I watch and/or follow sports. I feel that other sports media personalities are like family members to me. Watching and/or following my favorite sports media makes me feel like part of the sports media family. Sports media provide a behind-the-scenes view of teams and players to me. I like when sports media shows me the inner workings and lives of teams and players. I enjoy how sports media give me an all-access look at my favorite players and teams. Feelings of Sports Media Involvement (1=Strongly Disagree; 7= Strongly Agree) It is fun to watch/follow sports media. Watching/following sports media are some of the most satisfying things in my life. I really enjoy watching/following sports media. A lot of my life is organized around watching and/or following sports media. 56 (Beaton et. al, 2011) Table 2 (cont’d) SP_INV_M_C2 SP_INV_M_C3 Symbolic Value SP_INV_M_SV1 SP_INV_M_SV2 SP_INV_M_SV3 Unidimensional Sports Media Involvement (SP_INV_U) 5 SP_INV_U Items SP_INV_U1 SP_INV_U2 SP_INV_U3 SP_INV_U4 Watching and/or following sports media plays a central role in my life. I enjoy discussing what I read/see/hear on sports media with others. Watching and/or following sports media says a lot about who I am. Watching and/or following sports media tells others about me as a person. My habits of watching/following sports media gives a glimpse of the type of person I am. Feelings of Sports Media Involvement (1=Strongly Disagree; 7= Strongly Agree) A lot of my life is organized around watching and/or following sports media. Watching and/or following sports media plays a central role in my life. Watching and/or following sports media says a lot about who I am. Watching and/or following sports media tells others about me as a person. 57 (via Beaton et. al, 2011) Table 2 (cont’d) SP_INV_U5 Total Sports Media Consumption per Day (Total Consumption) 5 items Television Newspapers & Magazines Internet (for Sports News, Scores & Other Sports Media) Radio or Podcasts My habits of watching/following sports media gives a glimpse of the type of person I am. Hours of such type of Self-Reported sports media consumer Dependent Variable per day (None time spent consuming; Less than 1 hour; 1-2 hours; 2-3 hours; 3-4 hours; etc.) In an average day, I spend this amount of time watching or following sports on Television… In an average day, I spend this amount of time reading about sports in newspapers or magazines… In an average day, I spend this amount of time watching or following sports news, scores, YouTube videos, and other sports media on my laptop, Smartphone or portable device… In an average day, I spend this amount of time listening to sports talk radio and/or sports podcasts (on the actual radio or using a Smartphone app)… 58 Table 2 (cont’d) Social Media Gender 1 item In an average day, I spend this amount of time following and/or discussing sports on social media (e.g. Twitter, Facebook, Snapchat, Google+, etc.)… What is your Gender? Self-Reported Male or Female? Categorical Moderator Variable 59 CHAPTER 3 METHOD Participants A questionnaire of 72 items was administered to participants in order to analyze roughly 400 cases after data cleaning. Data from some participants were eliminated to prevent an overall biased sample due to evidence that some individuals had not read the question properly, were clicking on the exact same selection for each item, or were going too quickly to think about all the answers. Outliers who reported hours of sports media consumed per day more than three standard deviations from the mean were also removed from the sample (Malthouse, 2001). Survey items for the motivational factors were presented in random order to reduce any bias and a manipulation check question was placed in the survey randomly to determine if participants were completing the questionnaire properly. In order to acquire a representative sample of a subgroup within adults aged 18 and over in the United States who regularly consume sports media, data was collected via Amazon’s Mechanical Turk, which has been shown in academic research to be a reliable, inexpensive, and efficient avenue through which to collect valid data (Berinsky, Huber, & Lenz, 2012; Buhrmester, Kwang, & Gosling, 2011; Casler, Bickel, & Hackett, 2013). The sample came from a wide range of American adults and was representative of the sampling frame they create. After finishing the 20-minute survey, individuals were asked to submit a randomly generated survey completion code to validate their participation and receive the $0.76 incentive. Due to the fact that this project deals with sports media consumption, prospective participants were given a screener question prior to the survey about sports media usage: “Do 60 you regularly watch sports on television or follow sports online or via social media?” If an individual responded “No,” he or she was prohibited from participating in this study. Procedure The models analyzed in this research project consisted of the latent independent variables of both multidimensional and unidimensional team identity measures (tested separately), 12 separate antecedent motivations, and both multidimensional and unidimensional measures for sports media involvement (tested separately) as predictors of the observed dependent variable of total sports media consumption per day. Multidimensional team identity was measured via six separate 3-item scales, one scale for each first-order facet (Heere & James, 2007b; Heere et al., 2011), and unidimensional team identity was measured with a 4-item scale based on the Sports Spectator Identification Scale or SSIS (Wann & Branscombe, 1993). The motivational antecedent variables consist of 12 separate 3-item scales based on 12 discrete motivations to attend a live sporting event (Funk et al., 2004). Multidimensional sports media involvement was measured via three 3-item scales, one scale for each first-order facet (Beaton et al., 2011), and unidimensional sports media involvement was measured with a 5-item scale that exhibited construct validity using items from the three 3-item multidimensional first-order scales. Each latent variable item was measured on a 7-point Likert Scale ranging from Strongly Disagree (1) to Strongly Agree (7). The dependent variable of Total Sports Media Consumed per Day is an observed variable consisting of a sum of 4 items of self-reported hours spent consuming sports media per day on various media: Television, Internet, Radio, Newspapers and Magazines, and Social Media. Sports media consumption was measured in increments of one hour, beginning with the choice of zero hours consumed and ranging up to 12 hours per day for each type of media. 61 The antecedent motivational factors from Funk et al. (2004) were modified in order to examine the consumption of sports media. The previous 18 motivational variables from the 2004 study had three items each for a total of 54. The present study dropped six of the 18 original motivational variables, leaving a total of 36 items for 12 discrete Motivations to Consume Sports Media. The justification for dropping the six variables from the 2004 study (SWO – Support of women’s athletic events; ENT – Monetary value of attending a sporting event; WHO – Wholesome environment at a sporting event; STY – Style of play at a sporting event; MGT – Customer service at a sporting event; and BAS – Watching basketball live at a sporting event) is that these factors do not apply to consuming sports media. See Table 2 above for a summary of operationalized constructs and corresponding items for each antecedent motivational variable. A questionnaire of the aforementioned 72 items was completed by 400 participants in order to analyze data from 369 clean data rows. Statistical analysis by the software program Smart-PLS v.3.2.6 (Ringle, Wende, & Becker, 2015) employed Partial Least Squares Structural Equation Modeling (PLS-SEM) to calculate the path model on the mathematical basis of the path weighting scheme (Henseler, 2012). This program produces data to measure the validity of the constructs in the measurement model and the correlations between variables and variance explained by the structural models, as well as the significance levels of those correlations. Measurement of the results and the evaluation of the output of Smart PLS v.3.2.6 in the results section below were constructed following specific guidelines from the most prominent and most recent research regarding PLS-SEM and its applications (Chin, 2010; Hair, Ringle, & Sarstedt, 2011, 2012, 2013; Hair, Sarstedt, Pieper, & Ringle, 2012; Hair, Sarstedt, Ringle, & Mena, 2012). Through data gathered from a representative sample within a subgroup of adults aged 18 and over in the United States, this study aimed determine the effect of team identity, antecedent 62 motivations, and sports media involvement on total sports media consumption per day. It was expected that team identity would be mediated by the motivational factors and sports media involvement, and that link between motivational factors and consumption would be mediated by sports media involvement (Baron & Kenny, 1986). Due to the fact that statistical software like Smart-PLS v.3.2.6 (Ringle et al., 2015) can compute small sample sizes to adequately estimate relationships among variables, the participant number of N=369 is more than sufficient to allow for PLS-SEM analysis of the measurement and structural models in this project (Lowry & Gaskin, 2014; MacCallum, Widaman, Zhang, & Hong, 1999; Myers, Ahn, & Jin, 2011). The Use of Partial Least Squares Structural Equation Modeling (PLS-SEM) Partial Least Squares Structural Equation Modeling or PLS-SEM is a multivariate analysis method based on the principles of ordinary least squares multiple regression that seeks to maximize explained variance in dependent constructs (Chin, 1998, 2010; Hair et al., 2011; Hair, Ringle, et al., 2012; Vinzi, Chin, Henseler, & Wang, 2010). The approach differs from multiple regression in that it enables the simultaneous testing of the effects of latent variables on a criterion variable while also providing an evaluation of the data quality (Lowry & Gaskin, 2014). While there has been a steady increase of the use of PLS-SEM in management information systems (MIS) research in recent years (Ringle, Sarstedt, & Straub, 2012), a bulk of the published articles in the behavioral sciences has employed Covariance-Based Structural Equation Modeling or CB-SEM (Reinartz, Haenlein, & Henseler, 2009). CB-SEM is primarily used to reproduce a theoretical covariance matrix instead of explaining the variance of dependent constructs (Hair et al., 2011; Joreskog & Goldberger, 1975). The two techniques differ in how they approach parameter specification (Reinartz et al., 2009). Models employing CB-SEM 63 provide a measure of goodness of fit – the GFI or Goodness of Fit Index – as a way of confirming, via Confirmatory Factor Analysis, that the implied covariance structure of the model is suitably recreated by the sample data (Cheung & Rensvold, 2002). In other words, CB-SEM tests to see how well the data fits pre-established theoretical parameters. On the other hand, PLS-SEM is useful for testing models that build incrementally on past exploratory studies regarding prediction and the explanation of variance of dependent variables (Chin, 2010; Hair et al., 2013). PLS-SEM has seen an increase of usage in recent years in fields such as marketing research (Vinzi et al., 2010). It was created in the late 1960’s in order to evaluate path models and prediction models (Wold, 1974), and many recent studies point to this predictive ability as a justification of the method instead of CB-SEM (Rigdon, 2014). The originator of PLS-SEM wrote in one of the earliest publications on the topic, “Clearly, the predictor specification [of PLS-SEM] provides an immediate rationale for the operative use of the model for the purposes of prediction” (Wold, H., 1974, p.70). Current literature points out that CB-SEM is optimal for parameter accuracy while PLS-SEM is optimal for prediction accuracy (Sarstedt, Ringle, & Hair, 2014). Additionally, PLS-SEM is said to be more appropriate for exploratory research and theory development than for testing well-established theoretical models (Hair, Sarstedt, Ringle, et al., 2012). The use of this method, as opposed to CB-SEM, is best applied to studies such as the current one. While the previous chapters discussed established theories on both SIT and Involvement, as well as the motivational antecedents to Involvement, the combination of these theoretical constructs in a structural model has yet to be tested. The current project combines theories in an attempt to predict changes in total sports media consumed per day. A model significantly linking team identity and motivations to consume to time spent consuming media 64 has been successfully carried out by this author (Krier, 2016), but the current study incrementally extends this by adding the latent variable of sports media involvement. Data Analysis Questionnaire data relied on PLS-SEM to determine predictions of changes in amounts of sports media consumption in regards to the exogenous latent variables. These analyses were accomplished by the computer software program Smart-PLS v.3.2.6 (Ringle et al., 2015). SmartPLS allowed the research questions to be answered to determine if any of the 12 motivational factors have significant effects on the dependent variable, while also checking for differences in gender. SPSS v.22 (IBM Corp., 2017) was also used to perform separate OLS regressions where sports media involvement and total consumption were dependent variables, in order to doublecheck the data. The results of the OLS regression outputs were analyzed to ensure that they adhered to the assumptions of regression such as if there is indeed a linear relationship, if there is multivariate normality, and finally, if there is no multicollinearity, auto-correlation, nor homoscedasticity. Outliers that are outside of three standard deviations were dropped from the data set, but those that were deemed to be within 1 level of Cooke’s distance are acceptable to the data set and did not adversely affect the data analysis (Malthouse, 2001; Stevens, 1984). Regarding sample size, the desired N=369 sufficed for this project, as an increase in sample size helps to increase the power of the regression (Dupont & Plummer Jr., 1998), a statistical area that will not be considered in this project. The data output of Smart-PLS v.3.2.6 reported outer loadings (EFA results) of specific items for construct validity, cross-loadings for discriminant validity, AVE or Average Variance Extracted for convergent validity, Latent Variable Scores (weighted scores for each participant) 65 for determining the amount of variance explained for the different variables, and Chronbach’s Alpha levels of each construct for composite reliability. 66 CHAPTER 4 RESULTS Explanation of the Use of Unidimensional Scale Data Displayed in this Chapter Please note that the data displayed in this chapter reproduces the structural equation model using the unidimensional scales of both team identity and sports media involvement (Model #2; see Figure 2). Model #1, which tested the multidimensional scales of team identity and sports media involvement (see Figure 1), did not deliver a substantial difference in variance explained of the dependent variable of total sports media consumption per day over the more parsimonious Model #2 using unidimensional scales. More importantly, the multidimensional scales did not demonstrate discriminant validity from the other constructs in the model, so it was not clear if certain items were measuring more than one latent variable. Furthermore, each of the first-order facets of both team identity and sports media involvement exerted a nearly-equal amount of influence on each of the corresponding second-order variables. While the results shown in this chapter regarding Model #2 (using the unidimensional scales of team identity and sports media involvement) are significant and offer great insight into the psychological path influencing sports media consumption, the conclusion that the multidimensional scales are superfluous is still important to consider so as to inform researchers that the complexity of these constructs does not offer many, if any practical added benefits. Data results regarding Model #1 will be discussed near the end of this chapter, and a more thorough examination of this matter of the unnecessary nature of the multidimensional scales tested in Model #1 will take place in the discussion section. 67 Measurement Model Reflective measurement models, such as the ones tested in this study, are required to adhere to accepted standards of internal consistency (indicator and consistent reliabilities), as well as both convergent and discriminant validity (Klarner, Sarstedt, Hoeck, & Ringle, 2013). This section will provide the data to illustrate the reliability and validity of the constructs measured, along with descriptive statistics of each item. To first determine the convergent and discriminant validity of each construct examined in this research project, a factor analysis was conducted using the statistical software program SPSS v.22. Following the guidelines of the foremost literature on the subject (Fabrigar, Wegener, MacCallum, & Strahan, 1999), the factor analysis was run with the extraction method of maximum likelihood estimation using Varimax rotation, based on Eigenvalues greater than 1. The cutoff for a single factor was 0.6, and no cross-loadings were found within a 0.20 range for any items. Five latent variables that exhibited convergent and discriminant validity were revealed by the dimension reduction procedure. The five factors are, listed in order of percentage of overall variance, are as follows: unidimensional sports media involvement (SP_INV_U), unidimensional team identity (TI_U), Escape from Problems (ESC), Social Media Interaction (SOC), and Vicarious Achievement (VIC). Nine of the other antecedent motivational variables (PLA – Interest in Players, BON – Bonding with Friends, DRA – Sports Media Updates, COM – Community Connectedness, ROL – Role Models, EXC – Exciting Coverage, KNO – Knowledge of Sports, FAM – Family Atmosphere, and TEM – Team Interest Behind the Scenes) did not load onto a single factor, so each was dropped from the research model. Finally, since the 68 variable of total sports media consumption per day is a sum of five other measures, it becomes a single-item dependent variable and does not apply to any factor analysis examinations. Each item of a specific scale loaded at over 0.6 onto one factor, and did not cross-load in any meaningful way with any items from other constructs (all differences > 0.2). Along with the Average Variance Extracted (AVE) and Heterotrait-Monotrait Matrix (HTMT) tables that will be listed shortly, these data provide clear evidence of both convergent and discriminant validity for each of the constructs tested in this study. Table 3 shows outer-model loadings and crossloadings of the constructs along with Eigenvalues and percent of variance explained by each factor. Note than no single factor explains more than 50%, which indicates that each construct plays an important role in the model. The Bartlett’s test of Sphericity for the EFA presented in Table 3 was x2 = 5225.659 with df =153, p < .001, and the Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy was .910, which is over the .60 standard (H. Kaiser, Hunka, & Bianchini, 1971; H. F. Kaiser, 1958). 69 Table 3: Outer-Model Loadings and Cross-Loadings Component 3 .228 .282 .150 .171 .180 .072 .079 .158 .082 .861 .790 .843 .172 .132 .112 .109 .269 .191 1.532 8.51 Construct SP_INV_U1 SP_INV_U2 SP_INV_U3 SP_INV_U4 SP_INV_U5 TI_U_1 TI_U_2 TI_U_3 TI_U_4 ESC1 ESC2 ESC3 SOC1 SOC2 SOC3 VIC1 VIC2 VIC3 Eigenvalues % of Variance 1 .710 .653 .874 .862 .855 .155 .095 .215 .238 .169 .276 .255 .179 .175 .159 .295 .239 .261 8.309 46.16 2 .192 .297 .175 .136 .161 .836 .905 .822 .859 .128 .123 .090 .218 .207 .149 .164 .144 .083 2.201 12.23 4 .276 .117 .143 .137 .108 .127 .156 .200 .147 .100 .211 .134 .792 .829 .868 .215 .158 .163 1.325 7.36 5 .099 .273 .238 .242 .234 .184 .104 -.021 .152 .105 .253 .213 .213 .158 .136 .755 .803 .737 1.035 5.75 Extraction Method: Maximum Likelihood Estimation. Rotation Method: Varimax with Kaiser Normalization. (SP_INV_U: Sports Media Involvement; TI_U: Team Identity; ESC: Escape from Problems; SOC: Social Media Interaction; VIC: Vicarious Achievement.) 70 Table 4 provides the descriptive statistics and the reliability values of each construct. Table 4: Descriptive Statistics and Reliability of Constructs Construct Item Mean Std. Dev. Loading Weight tStatistic Composite reliability Cronbach's α AVE SP_INV_U1 3.57 1.667 0.827 42.868 0.946 0.929 0.780 SP_INV_U2 3.83 1.609 0.839 47.188 SP_INV_U3 3.71 1.576 0.927 112.835 SP_INV_U4 3.76 1.559 0.911 84.225 SP_INV_U5 3.91 1.564 0.906 75.304 TI_U1 5.51 1.238 0.877 52.442 0.945 0.922 0.810 TI_U2 5.61 1.302 0.920 79.818 TI_U3 5.31 1.531 0.879 44.050 TI_U4 5.21 1.431 0.923 101.078 ESC1 4.69 1.601 0.869 45.734 0.931 0.890 0.819 ESC2 4.19 1.627 0.922 101.046 ESC3 4.35 1.652 0.923 105.122 SOC1 4.48 1.597 0.896 62.563 0.898 0.828 0.745 SOC2 4.16 1.716 0.898 75.963 SOC3 4.33 1.697 0.896 56.808 VIC1 3.83 1.687 0.868 58.245 0.925 0.879 0.804 VIC2 3.98 1.634 0.898 68.428 VIC3 3.86 1.526 0.822 36.204 Unidimensional Sports Media Involvement (SP_INV_U) Unidimensional Team Identity (TI_U) Escape from Problems (ESC) Social Media Interaction (SOC) Vicarious Achievement (VIC) 71 Each construct exhibits composite reliability and a Chronbach’s α level above 0.80, which is the threshold for good internal consistency. Each construct also has an AVE value of over 0.50, also illustrating its discriminant validity (Klarner et al., 2013). Furthermore, each item also has a normal distribution. They were tested for skewness and kurtosis, and no item had a skewness of over +/- 2.0, nor did any exhibit kurtosis of over +/- 7. Regarding collinearity statistics, no two variables exhibited multicollinearity, shown by the variable inflation factor measures (VIF) in Table 5. All of the VIF values are all under the threshold of 2.5, illustrating no existence of multicollinearity. Table 5: Variable Inflation Factor (VIF) – Multicollinearity Statistics 1.193 TI_U / SP_INV_U 1.418 ESC/ SP_INV_U 1.468 1.276 VIC / SP_INV_U SOC / TOTAL_CONSUMPTION 1.276 SP_INV_U / TOTAL_CONSUMPTION (VIF < 2.5 indicates no multicollinearity). (SP_INV_U: Sports Media Involvement; TI_U: Team Identity; ESC: Escape from Problems; SOC: Social Media Interaction; VIC: Vicarious Achievement.) 72 Table 6 illustrates the results of a correlation analysis of the model, which was run using SPSS v.22. Average variance extracted (AVE), produced by Smart-PLS v.3.2.6 is shown in bold on the diagonal of Table 6. The square root of AVE for each is recommended to be higher than the correlations between any two constructs (Fornell & Larcker, 1981; Shook, Ketchen Jr., Hult, & Kacmar, 2004). This is the case for each construct tested in the research model. Table 6: Correlations of Constructs & Discriminant Validity Assessment SP_INV_U SP_INV_U TI_U ESC SOC VIC Total Consumption 0.883 TI_U .468 0.900 ESC .548 .326 0.905 SOC .463 .442 .409 0.897 VIC .614 .367 .518 .480 0.863 TOT_CONS .436 .297 .262 .348 .303 1.000 *** NOTE: All correlations are significant at the 0.001 level (2-tailed) Diagonal elements are the square root of AVE or √AVE, and non-diagonal elements are correlations between latent variables (Fornell-Larcker Criterion of Discriminant Validity) (Fornell & Larcker, 1981) (SP_INV_U: Sports Media Involvement; TI_U: Team Identity; ESC: Escape from Problems; SOC: Social Media Interaction; VIC: Vicarious Achievement; TOT_CONS: Total Sports Media Consumption per Day.) 73 The final check of discriminant validity for the constructs of the measurements model is shown in Table 7, which is the Heterotrait-Monotrait Matrix (HTMT). Discriminant validity is shown for values lower than the threshold of 0.850. As can be seen in Table 7, each of the constructs exhibits discriminant validity accordion to the HTMT criterion (Henseler, Ringle, & Sarstedt, 2015). Table 7: Heterotrait-Monotrait (HTMT) Matrix of Discriminant Validity ESC SOC SP_INV_U TI_U TOT_CONS VIC ESC SOC SP_INV_U TI_U TOT_CONS VIC 0.463 0.602 0.358 0.278 0.604 0.512 0.490 0.372 0.563 0.502 0.450 0.700 0.309 0.420 0.332 - (HTMT values < 0.850 illustrate discriminant validity). (SP_INV_U: Sports Media Involvement; TI_U: Team Identity; ESC: Escape from Problems; SOC: Social Media Interaction; VIC: Vicarious Achievement; TOT_CONS: Total Sports Media Consumption per Day.) To conclude, the assessment of the measurement model demonstrates that all of the constructs measured in this study are both fully reliable and valid. Since this has now been established, an evaluation of the structural model (Model #2) can take place, along with an analysis of the hypotheses and research questions testing the relationships between the variables. 74 Structural Model The structural model linking unidimensional team identity, three antecedent motivations (Escape from Problems – ESC, Social Media Interaction – SOC, and Vicarious Achievement – VIC), unidimensional sports media involvement, and total sports media consumption per day will now be statistically analyzed in a step-by-step fashion in order to present a comprehensive view of the results. This will allow for a test of hypotheses and provide answers to each research question regarding mediation. First, the research model will be illustrated in its simplest form in Figure 3, with only significant pathways shown for ease of comprehension. This will allow for a clear understanding of the relationship between the variables discussed in the measurement model section above which were shown to be both reliable and valid. Next, the step-by-step analysis will begin with the direct relationship between team identity and total sports media consumption per day. Building from that starting point, additional variables (sports media involvement, then the antecedent motivations) will be added to the PLS structural model to illustrate mediation effect as per the general recommendations by Baron and Kenny (1986). The complete structural model will then be illustrated and discussed in Figure 12, with data tables shown for the predictive path coefficients (including indirect, direct and total effects) and variance explained of the criterion variables of unidimensional sports media involvement and total sports media consumption. These data will enable an interpretation of the path coefficients as predicted increases in minutes of sports media consumed per day. Finally, a comparison of the structural model with regards to gender used as a categorical moderator variable for multi-group analysis will be discussed (Gefen & Straub, 1997; Rigdon, Ringle, & Sarstedt, 2010; Sarstedt, 2008; Sarstedt, Henseler, & Ringle, 2011). This comparison 75 will illustrate significant differences in the ways in which sports media are consumed by different cultures of male and female sports fans. To begin the analysis of the models, Figure 3 below illustrates the unidimensional structural model tested in this project, showing only significant connections between variables (the complete model with nonsignificant paths as red lines will be shown at the culmination of this analysis in Figure 12). Path coefficients and adjusted R2 data will be discussed as applicable. For ease of interpretation, the following color scheme is assigned to the constructs in the structural model: Team Identity (TI_U) – Yellow; Antecedent Motivations (ESC, VIC, & SOC) – Red; Sports Media Involvement (SP_INV_U) – Green; Total Sports Media Consumption per Day (TOT_CONSUM) – Blue. 76 Figure 3: Complete Structural Model of Unidimensional Constructs Complete Structural Model of Unidimensional Team Identity and Unidimensional Sports Media Involvement as predictors of Total Sports Media Consumption per Day. NOTE: Only Significant paths pictured. (TI_U: Team Identity; ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction; SP_INV_U: Sports Media Involvement; TOT_CONSUM: Total Sports Media Consumption per Day.) The software program Smart-PLS v.3.2.6 was used for the analysis of the structural model. PLS accounts for errors in measurement of each of the observed variables, and is preferable to ordinary least squares (OLS) regression due to this element of PLS-SEM (Chin, 1998; Hair et al., 2011; J. Osborne & Waters, 2002). As recommended by the leading literature in the area of PLS-SEM, bootstrapping was set for 5000 samples taken from the data set of 369 77 subjects, and two-tailed significance at the 95% level was determined if path coefficients produced were accompanied by t-statistics over the cut-off point of 1.96 (Hair, Hult, Ringle, & Sarstedt, 2017; Vinzi et al., 2010). The critical value cut-off point for a two-tailed 99% level of significance is a t-statistic over 2.58, and the critical value cut-off point for a two-tailed 99.9% level of significance is a t-statistic over 3.30 (Dunnett, 1964). To test H1, a model consisting of only team identity and total sports media consumption was computed for the amount of direct influence team identity has on consumption. The results are shown in Figure 4. Figure 4: Team Identity Direct to Total Sports Media Consumption per Day Direct Path of Unidimensional Team Identity to Total Sports Media Consumption per Day. The path coefficient is shown along arrow with t-statistic in parentheses. Adjusted R2 value shown inside circle. *** p < 0.001. (TI_U: Team Identity; TOT_CONSUM: Total Sports Media Consumption per Day.) Figure 4 illustrates that a one-unit increase in TI_U results in a standardized coefficient increase of 0.299 of Total Consumption. This path coefficient has a t-statistic of 6.970 (p < 0.001). Thus, H1 is supported. Higher levels of team identity result in higher levels of total sports media consumption per day. The single predictor variable explains an adjusted R2 of 8.7%, which is low amount of variance explained of the endogenous construct of total consumption. This simple model will serve as a basis for the remainder of the analysis that follows. To test H2 regarding team identity and sports media involvement, another direct path model was run. Figure 5 shows the results of that model. 78 Figure 5: Team Identity Direct to Unidimensional Sports Media Involvement Direct Path of Unidimensional Team Identity to Unidimensional Sports Media Involvement. The path coefficient is shown along arrow with t-statistic in parentheses. Adjusted R2 value shown inside circle. *** p < 0.001. (TI_U: Team Identity; SP_INV_U: Sports Media Involvement.) Figure 5 illustrates that a one-unit increase in TI_U results in a standardized coefficient increase of 0.474 of SP_INV_U. This path coefficient has a t-statistic of 12.758 (p < 0.001). Thus, H2 is supported. Higher levels of team identity result in higher levels of sports media involvement. The single predictor variable explains an adjusted R2 of 22.2%, which is moderately low amount of variance explained of the endogenous construct of sports media involvement. Since we have now inspected the direct paths from team identity to both total consumption and sports media involvement, we will begin to assess the possible existence of mediation by combining constructs in the measurement model as directed by theory (Funk et al., 2004; Tajfel, 1974; Zaichkowsky, 1985). This examination will begin with H7, which hypothesizes that sports media involvement mediates the relationship between team identity and total consumption. Mediation is existent when there is a drop in the path coefficient and the significance of the direct path between two constructs in the presence of an intermediate construct while indirect paths are each individually significant (Baron & Kenny, 1986; 79 MacKinnon et al., 2006). This relationship between team identity and total consumption with sports media involvement acting as a mediator is illustrated in Figure 6 below. Figure 6: Sports Media Involvement Mediating Team Identity to Total Consumption Model showing Unidimensional Sports Media Involvement mediating the relationship between Unidimensional Team Identity and Total Sports Media Consumption per Day. The path coefficients are shown along arrows with corresponding t-statistics in parentheses. Adjusted R2 values shown inside circles. *** p < 0.001; * p < 0.05. (TI_U: Team Identity; SP_INV_U: Sports Media Involvement; TOT_CONSUM: Total Sports Media Consumption per Day.) For H7 regarding the mediation of the direct path from TI_U to Total Consumption, the results shown in Figure 6 reveal that SP_INV_U did mediate the relationship. As seen previously in Figure 4, a model involving only Team Identity as a predictor of Total Consumption, the direct path coefficient was 0.299 with a t-statistic of 6.970 (p < 0.001). When incorporating the construct of unidimensional sports media involvement into the model, the direct effect from TI_U to Total Consumption dropped to 0.115 (Δ = -0.184) and the significance level was reduced to a t-value of 2.184 (p < 0.05). Because the path from TI_U to SP_INV_U was 80 statistically significant and the paths from SP_INV_U to Total Consumption was also significant, this drop in the effect of TI_U to Total Consumption indicated partial mediation (Baron & Kenny, 1986). Thus, H7 is supported. Additionally, because the path coefficient from sports media involvement to total consumption reveals that a one-unit increase in SP_INV_U predicts a 0.387 increase in total consumption with a t-statistic of 8.615 (p < 0.001), this also supports H6 regarding the direct effect of increased sports media involvement predicting an increase in total consumption. Finally, note that the adjusted R2 of total consumption increases from 8.7% to 20.1% (Δ = 12.4%) due to the inclusion of SP_INV_U as a mediating variable. This is now a moderately low but satisfactory amount of variance explained in the criterion variable of total consumption. Continuing with the mediation analysis, H8 relates to whether the antecedent motivations discussed in the measurement model section (ESC – Escape from Problems, VIC – Vicarious Achievement, and SOC – Social Media Interaction) mediate the relationship between team identity and sports media involvement. H9 relates to whether the antecedent motivations mediate the relationship between team identity and total consumption. Figure 7 displays the model testing H8, and Figure 8 displays the model testing H9. 81 Figure 7: Antecedent Motivations Mediating Team Identity to Sports Involvement Model showing three Antecedent Motivations mediating the relationship between Unidimensional Team Identity and Unidimensional Sports Media Involvement. The path coefficients are shown along arrows with corresponding t-statistics in parentheses. Adjusted R2 value shown inside circles. *** p < 0.001. (TI_U: Team Identity; ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction; SP_INV_U: Sports Media Involvement.) For H8 regarding the mediation of the direct path from TI_U to SP_INV_U by antecedent motivations, the results shown in Figure 7 reveal that the three antecedent motivations do not mediate the relationship according to the specifications of Baron & Kenny (1986). The reasoning is seen in the path from SOC to SP_INV_U, which is not significant, as paths from any mediator to a dependent variable must show significance in order to claim mediation (MacKinnon et al., 2006). We will see later in Figure 10 that the direct path from SOC to SP_INV_U with no antecedents does produce a significant relationship, which indicates a strong effect of team 82 identity on sports media involvement that lessens the effect of SOC. Nonetheless, Figure 7 does not illustrate mediation. These findings make logical sense, as the antecedents relate to constructs that follow them in a structural pathway not necessarily to what might be antecedent to them. Though the inclusion of the antecedent motivations changes the direct effect from team identity to sports media involvement seen in Figure 5 (Δ = -0.261) with significance level reduced to a t-value of 4.618 (p < 0.01) from 12.758, mediation cannot be substantiated. Therefore, H8 is not supported. It should be noted that the adjusted R2 of sports media involvement increases from 22.2% to 49.9% (Δ = 27.7%) due to the inclusion of the significant antecedent paths from ESC and VIC (compare Figure 5 and Figure 7). So while H8 is not supported in the context of all three antecedent motivations, it is clear that the presence of ESC and VIC along with TI_U explain a moderately high and satisfactory amount of variance in the construct of unidimensional sports media involvement seen above in Figure 7. 83 Figure 8: Antecedent Motivations Mediating Team Identity to Total Consumption Model showing three Antecedent Motivations mediating the relationship between Unidimensional Team Identity and Total Sports Media Consumption per Day. The path coefficients are shown along arrows with corresponding t-statistics in parentheses. Adjusted R2 value shown inside circles. *** p < 0.001; ** p < 0.01; * p < 0.05. (TI_U: Team Identity; ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction; TOT_CONSUM: Total Sports Media Consumption per Day.) Following the reasoning presented in the preceding paragraph, for H9 regarding the mediation of the direct path from TI_U to Total Consumption, the results shown in Figure 8 reveal that the three antecedent motivations again do not mediate the relationship according to the recommendations of Baron & Kenny (1986). The reasoning is seen in the path from ESC to TOT_CONSUM, which is not significant, as paths from any mediator to a dependent variable must show significance in order to claim mediation (MacKinnon et al., 2006). This is consistent with the findings shown later in Figure 11 below where the direct path from ESC to total consumption with no antecedents does not produce a significant relationship. There is a change 84 in the direct effect from TI_U to Total Consumption of -0.158 and the significance level was reduced to a t-value of 2.542 (p < 0.01) from 6.970. This reveals an effect on the relationship between team identity and total consumption due to VIC and SOC, while ESC has no effect on the relationship. Nonetheless, H9 is not supported. Finally, the adjusted R2 of total consumption increases from 8.7% to 15.9% (Δ = 7.2%) due to the inclusion of VIC and SOC (comparing Figure 4 to Figure 8). This is a moderately low amount of variance explained. Next, although the direct paths from TI_U to the antecedent motivations were shown in Figures 7 and 8, H3 will now be examined to continue this step-by-step inspection of the structural model. Figure 9 illustrates the relationships between the constructs of TI_U, ESC, VIC, and SOC. Figure 9: Direct Paths of Team Identity to Antecedent Motivations Direct Paths of Unidimensional Team Identity to Antecedent Motivations. The path coefficients are shown along arrows with corresponding t-statistics in parentheses. Adjusted R2 value shown inside circles. *** p < 0.001. (TI_U: Team Identity; ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction.) 85 The results of the direct paths between TI_U and the antecedent motivations reveal that a one-unit increase in TI_U results in a standardized coefficient increase of 0.327 of ESC with a tstatistic of 6.461 (p < 0.001), an increase of 0.372 of VIC with a t-statistic of 8.444 (p < 0.001), and an increase of 0.443 of SOC with a t-statistic of 10.075 (p < 0.001). Thus, H3 is supported. Higher levels of team identity result in higher levels of each of the antecedent motivations. The predictor variable explains an adjusted R2 of 10.4% of ESC, 13.6% of VIC, and 19.4% of SOC, which are each moderately low amounts of variance explained of the antecedent motivation constructs. The next hypothesis, H4, posits that the antecedent motivations will each predict significant increases in sports media involvement. Figure 10 illustrates the direct relationship between three antecedent motivations (ESC, VIC, and SOC) and the construct of unidimensional sports media involvement. 86 Figure 10: Direct Paths of Antecedent Motivations to Sports Media Involvement Direct Paths of Antecedent Motivations to Unidimensional Sports Media Involvement. All paths significant at p < 0.001. The path coefficients are shown along arrows with corresponding tstatistics in parentheses. Adjusted R2 value is shown inside the green circle. *** p < 0.001; ** p < 0.01. (ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction; SP_INV_U: Sports Media Involvement.) The results of the direct paths from the antecedent motivations to sports media involvement in Figure 8 reveal that a one-unit increase in ESC results in a standardized coefficient increase of 0.283 of SP_INV_U with a t-statistic of 5.288 (p < 0.001), a one-unit increase in VIC results in a standardized coefficient increase of 0.391 of SP_INV_U with a tstatistic of 6.985 (p < 0.001), and a one-unit increase in SOC results in a standardized coefficient increase of 0.156 of SP_INV_U with a t-statistic of 2.977 (p < 0.01). Thus, H4 is supported. Higher levels of each antecedent motivation results in an increase in sports media involvement. 87 The predictor variables explains an adjusted R2 of 46.5% of SP_INV_U, which is a moderately high and satisfactory amount variance explained of the sports media involvement construct. Hypothesis H5 posits that the antecedent motivations will each predict significant increases in total sports media consumption per day. Figure 11 illustrates the direct relationship between three antecedent motivations (ESC, VIC, and SOC) and the construct of total consumption. Figure 11: Direct Paths of Antecedent Motivations to Total Consumption Direct Paths of Antecedent Motivations to Total Sports Media Consumption per Day. The path coefficients are shown along arrows with corresponding t-statistics in parentheses. Adjusted R2 value is shown inside the blue circle. *** p < 0.001; * p < 0.05. (ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction; TOT_CONSUM: Total Sports Media Consumption per Day.) The results of the direct paths from the antecedent motivations to total consumption in Figure 11 reveal that a one-unit increase in ESC results in a standardized coefficient increase of 0.090 of total consumption with a t-statistic of 1.552 (not significant at the 95% level), a one-unit 88 increase in VIC results in a standardized coefficient increase of 0.135 of total consumption with a t-statistic of 2.474 (p < 0.01), and a one-unit increase in SOC results in a standardized coefficient increase of 0.250 of total consumption with a t-statistic of 4.223 (p < 0.01). Thus, H5 is partially supported. Higher levels of two of the three antecedent motivations (VIC and SOC) results in an increase in total consumption. The path coefficient from ESC to total consumption is both small and insignificant. In total, the three predictor variables explain an adjusted R2 of 14.6% of total consumption, which is a moderately low amount variance explained of the sports media involvement construct by the antecedent motivations. Now that the components of the complete structural model have been laid out in a stepby-step sequence illustrating the sectional relationships, the entire model consisting of all six constructs can be examined as a whole. This will enable a final mediation analysis and provide a clear view of the relationships between the variables tested in this study. Figure 12 illustrates the complete structural model of team identity, antecedent motivations, and sports media involvement as predictors of total sports media consumption per day. Please note that Figure 12 uses red arrows for each of the nonsignificant relationships. The significant relationships are shown with black arrows, with green ovals around the significant path coefficients and t-statistics for ease of detection. For reference, an unobstructed view of the significant paths can be seen by referring back to Figure 3. The results of the structural model in Figure 12 are also displayed in Table 8 below. 89 Figure 12: Complete Structural Model Complete Structural Model with T-values in parentheses. Significant path coefficients are shown along black arrows circled by green ovals. Nonsignificant paths are shown along red arrows. Adjusted R2 values are shown inside circles. *** p < 0.001; ** p < 0.01. (TI_U: Team Identity; ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction; SP_INV_U: Sports Media Involvement; TOT_CONSUM: Total Sports Media Consumption per Day.) The structural model displayed in Figure 12 reveals each of the relationships between the variables of team identity, antecedent motivations, sports media involvement, and total sports media consumption per day. Note that path coefficients in Figure 12 may differ slightly from other similar relationships in Figures 4-9 due to the random sampling of the bootstrap procedure (Vinzi et al., 2010). 90 Starting with TI_U, the effects of team identity on the three antecedent motivations are the same as what was illustrated in Figure 9 (all are significant relationships). The effect of team identity on sports media involvement is significant with the same path coefficient value as illustrated in Figure 7 (0.213 coefficient; p < 0.001). The effects of the antecedent motivations of ESC and VIC are significant predictors of increases in sports media involvement (again similar values shown in Figure 7), and in conjunction with team identity, these three constructs produce an adjusted R2 of 50.0% for the explained variance of SP_INV_U in the structural model listed above. This is a moderately high amount of variance explained and illustrates the satisfactory explanatory ability of this research model. Next, the antecedent motivation of SOC was the only significant motivational predictor of total consumption. Figure 12 reveals that a one-unit increase in SOC results in a standardized coefficient increase of 0.173 of total consumption with a t-statistic of 2.778 (p < 0.01). SOC appears to be a contributing factor to the full mediation of team identity to total consumption, which will be addressed in more detail in a moment. The biggest direct influence on predicted increase in total sports media consumption per day was due to the construct of sports media involvement. A one-unit increase in SP_INV_U results in a standardized coefficient increase of 0.334 of total consumption with a t-statistic of 5.125 (p < 0.001). Overall, the exogenous constructs of team identity, three antecedent motivations (ESC, VIC, & SOC), and sports media involvement produce an adjusted R2 of 21.2% for the explained variance of total sports media consumption per day. This is a moderately low amount of variance explained, generally speaking. That said, since the sample was representative and only surveyed individuals who regularly watched or followed sports media on a daily basis, it can be expected that there is less variance overall to explain in the dependent 91 variable of time spent consuming sports media. Therefore, this amount of variance explained is satisfactory to substantiate the worth of this structural model. Regarding the mediation of the direct path between team identity and total consumption, Figure 12 reveals that ESC, VIC, SOC, & SP_INV_U fully mediate the relationship between TI_U and Total Consumption. The nonsignificant direct path shown above has a path coefficient of 0.071, which is a change of -0.228. Thus, a bulk of the effect of team identity on total consumption must travel through the other mediating variables to affect changes in time spent watching or following sports media each day. Three other nonsignificant paths are illustrated in Figure 12, all relating to the antecedent motivations. Escape from Problems (ESC) and Vicarious Achievement (VIC) are significant predictors of increased sports media involvement and have almost no direct effect on total consumption. Social Media Interaction (SOC) is a significant predictor of increased total consumption, but has no significant effect on sports media involvement at the 95% level. Lastly, regarding Figure 12, there is a measure of predictive relevance called Q2 that is used to indicate whether the structural model can accurately predict data not used in the sample taken for this study (Geisser, 1974; Klarner et al., 2013; Stone, 1974). Q2 is a value produced by SmartPLS v.3.2.6 by conducting a blindfolding procedure that randomly omits data points in the sample while running the PLS algorithm. Any resulting Q2 value larger than zero illustrates that the model has predictive relevance for each endogenous construct. This is the case with the current structural model (all Q2 > 0), and predictive relevance values can been seen in Table 8. It should be noted that sports media involvement produced the largest Q2 score, showing the strength of its overall predictive relevance (Chin, 1998). 92 Table 8 shows the full assessment of the PLS structural model illustrated in Figure 12. Note that per the recommendations of existing literature (Hair et al., 2017; Vinzi et al., 2010), the bias-corrected 95% confidence intervals for each path coefficient are listed along with the direct path coefficients and corresponding significance (Efron, 1987). Table 8: PLS Structural Model Assessment Endogenous constructs Sports Media Involvement Total Consumption ESC SOC VIC Relationship between constructs TI_U  ESC TI_U  VIC TI_U  SOC TI_U  SP_INV_U TI_U  TOT_CONS ESC  SP_INV_U ESC  TOT_CONS VIC  SP_INV_U VIC  TOT_CONS SOC  SP_INV_U SOC  TOT_CONS SP_INV_U  TOT_CONS Adjusted R2 0.500 0.212 0.104 0.135 0.195 Q2 0.365 0.197 0.081 0.148 0.097 Direct path coefficient p-value Bias-Corrected 95% confidence interval 0.327 0.371 0.444 0.213 0.071 0.260 -0.012 0.356 -0.006 0.090 0.173 0.334 0.000 0.000 0.000 0.000 0.198 0.000 0.833 0.000 0.920 0.074 0.005 0.000 [0.225; 0.423] [0.277; 0.459] [0.346; 0.532] [0.126; 0.308] [-0.039; 0.178] [0.156; 0.365] [-0.121; 0.099] [0.245; 0.459] [-0.123; 0.120] [-0.009; 0.186] [0.046; 0.290] [0.201; 0.457] Note: Q2 is derived from a blindfolding procedure with an omission distance of seven; the p-values are produced from a bootstrapping procedure with 5,000 samples of the 369 cases. The biased-corrected two-tailed 95% confidence intervals are based on the recommendations of existing research (Efron, 1987). (TI_U: Team Identity; ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction; SP_INV_U: Sports Media Involvement; TOT_CONS: Total Sports Media Consumption per Day.) 93 Comparing the model seen in Figure 12 to the other models presented in this section enables the determination of the full extent of mediation along the paths of the structural model. The values in Figure 12 reveal the direct effects only. In order to determine the differential effects of each exogenous construct on endogenous constructs, total effects are calculated by adding the direct effect to the product of the indirect effects (Hair et al., 2017). SmartPLS v.3.2.6 provides the significance levels for each of the individual total effects. The direct and indirect effects, as well as the total effects are expressed in Table 9. Table 9: Mediating Effects Analysis Relationship between constructs TI_U  TOT_CONS TI_U  SP_INV_U ESC  TOT_CONS VIC  TOT_CONS SOC  TOT_CONS Direct Effect Indirect Effect 0.071 0.213*** -0.012 -0.006 0.173** 0.228*** 0.257*** 0.087*** 0.119*** 0.030 Total Effect 0.299*** 0.470*** 0.075 0.113* 0.203** *** p < 0.001, ** p < 0.01, * p < 0.05. (TI_U: Team Identity; ESC: Escape from Problems; VIC: Vicarious Achievement; SOC: Social Media Interaction; SP_INV_U: Sports Media Involvement; TOT_CONS: Total Sports Media Consumption per Day.) The results shown in Table 9 indicate that team identity has a strong total effect on sports media consumption per day with a coefficient of 0.299 (p < 0.001). This is only second to the effect of sports media involvement on total consumption seen in Table 8 (0.334, p < 0.001). Team identity also has a large and significant total effect on sports media involvement, thanks in part to the mediation by the motivational variables of ESC and VIC. This indicates the overall importance of team identity in this model as the first step towards fostering more time spent consuming sports media. 94 The results of the hypotheses are shown in Table 10. Seven of the nine hypotheses were either supported or partially supported. Analysis of the research questions dealing with the multidimensional variables will be discussed after Table 10. H1: Table 10: Hypotheses Results Higher levels of Team Identity will result in higher levels of Total Sports Media Consumption per day. Supported H2: Higher levels of Team Identity will result in higher levels of Sports Media Involvement. Supported H3: Higher levels of Team Identity will result in higher levels of significant predictor Antecedent Motivations. Supported H4: Higher levels of significant predictor Supported Antecedent Motivations will result in higher levels of Sports Media Involvement. Higher levels of Antecedent Motivations Partially Supported will result in higher levels of Total Sports Media Consumption per day. H5: H6: H7: H8: H9: Higher levels of Sports Media Involvement will result in higher levels of Total Sports Media Consumption per day. Sports Media Involvement mediates the relationship between Team Identity level and Total Sports Media Consumption per Day. Antecedent Motivations mediate the relationship between Team Identity and Sports Media Involvement. Supported Antecedent Motivations mediate the relationship between Team Identity and Total Sports Media Consumption per Day. Not Supported 95 Supported Not Supported The first three research questions posed in an earlier section each dealt with the notion of multidimensional constructs of team identity and sports media involvement versus the unidimensional scales used in the structural models tested in the current chapter. Unfortunately, a direct comparison cannot be conducted because the multidimensional versions of both team identity and sports media involvement did not adhere to the rigorous measurement model standards of convergent and discriminant validity. Therefore, the battery of diagnostic statistical tests for investigating the relevance of a structural model would not have allowed for an acceptable source of information for model interpretation. As such, RQ1 (“Does Model #1 with Multidimensional Team Identity (Heere & James, 2007b) and Multidimensional Sports Media Involvement (Beaton et al., 2011) explain more variance of Total Sports Media Consumption per day than Model #2 using Unidimensional Team Identity (Wann & Branscombe, 1993) and Unidimensional Sports media involvement?”) is unable to be answered. The parsimonious nature of the unidimensional model (shown earlier in Figure 2) shows that a complete predictive model of team identity, antecedent motivations, and sports media involvement as predictors of total sports media per day is sufficient for producing significant and meaningful results. Regarding RQ2 (“Which first-order facets of Multidimensional Team Identity contribute most to the identity construct?”) and RQ3 (“Which first-order facets of Multidimensional Sports Media Involvement contribute most to the involvement construct?”), all six of the first-order facets of multidimensional team identity – public evaluation, private evaluation, interconnection of self with the team, sense of interdependence with the team, behavioral involvement, and cognitive awareness (Ashmore et al., 2004; Heere & James, 2007b; Heere et al., 2011) – shared a relatively even amount of influence on the second-order team identity level variable. The same is 96 true for the first-order facets of multidimensional sports media involvement – hedonic value, centrality to lifestyle, and symbolic value (Beaton et al., 2011). They too share a relatively even amount of influence on the second-order sports media involvement variable. Table 11 lists the first-order path coefficients and descriptive statistics for multidimensional team identity and multidimensional sports media involvement. Please see Table 11 at the top of the following page. 97 Table 11: First-Order Facets of Multidimensional Constructs Construct First-Order Facet Item Mean Std. Dev. Chronbach’s Influence on Second order construct t-statistic 0.903 0.221 20.328 0.861 0.170 15.557 0.901 0.247 22.127 0.838 0.230 23.706 0.761 0.183 24.594 0.885 0.184 17.685 α Public Evaluation TI_M_PbE1 4.39 1.637 TI_M_PbE2 4.05 1.668 TI_M_PbE3 4.66 1.499 TI_M_PvE1 6.02 1.076 Team Identity TI_M_PvE2 5.91 1.109 (TI_M) TI_M_PvE3 5.50 1.352 TI_M_IntD1 3.78 1.674 TI_M_IntD2 3.48 1.597 TI_M_IntD3 3.94 1.641 TI_M_IntC1 3.94 1.784 TI_M_IntC2 4.28 1.632 TI_M_IntC3 4.06 1.762 TI_M_B1 5.31 1.474 TI_M_B2 4.25 1.675 TI_M_B3 4.56 1.664 TI_M_CA1 5.37 1.467 TI_M_CA2 5.43 1.372 Private Evaluation Multidimensional Sense of Interdependence with the Group Interconnection with the Group Behavioral Involvement Cognitive Awareness 98 Table 11 (cont’d) Hedonic Value Multidimensional Sports Media Involvement Centrality to Lifestyle (SP_INV_M) TI_M_CA3 5.65 1.327 SP_INV_HV1 5.25 1.282 SP_INV_HV2 4.31 1.528 SP_INV_HV3 5.03 1.338 SP_INV_C1 3.57 1.667 SP_INV_C2 3.83 1.609 SP_INV_C3 4.42 1.555 SP_INV_SV1 3.71 1.576 SP_INV_SV2 3.76 1.559 SP_INV_SV3 3.91 1.564 0.887 0.358 38.601 0.811 0.354 42.263 0.954 0.407 41.109 Symbolic Value Placing focus back on the unidimensional structural model (Model #2), the final research question, R4 (“Are there significant gender differences in the motivational drivers of sports media involvement and sports media consumption?”) requires the variable of gender to be added to the model as a categorical moderator variable (Gefen & Straub, 1997; Henseler & Fassott, 2010; Rigdon et al., 2010; Sarstedt, 2008). SmartPLS v.3.2.6 allows for multi-group analysis (MGA) by separating the data into two groups before comparing path coefficients for significant differences via a bootstrapping procedure. This PLS-MGA analysis results in a data output similar to a t-test (Sarstedt et al., 2011). The sample of 369 individuals tested in this study consisted of 222 males (60.2%) and 147 females (39.8%), which is a similar proportion to the reported population of adult sports media consumers in the United States (Adweek, 2014). 99 The results of the PLS-MGA testing gender differences in the structural model linking unidimensional team identity, antecedent motivations (ESC, VIC, & SOC), and unidimensional sports media involvement as predictors of total sports media consumption per day are shown in Table 12. Table 13 shows the means and standard deviation for time spent consuming sports media for males and females in this study. Table 12: Gender Differences in Structural Model Relationships Relationship between constructs Gender Path Coefficient t-statistic p-value Path Coefficient |Difference| 0.208 p-value of gender Difference 0.024* ESC  SP_INV_U Male Female 0.323 0.115 5.020 1.418 0.000*** 0.156 VIC  SP_INV_U Male Female 0.276 0.549 4.148 7.793 0.000*** 0.000*** 0.273 0.003** TI_U  SP_INV_U Male Female 0.286 0.036 4.869 0.469 0.000*** 0.639 0.250 0.004** SP_INV_U  TOT_CONS Male Female 0.378 0.130 4.483 1.106 0.000*** 0.269 0.248 0.044* *** p < 0.001, ** p < 0.01, * p < 0.05. (TI_U: Team Identity; ESC: Escape from Problems; VIC: Vicarious Achievement; SP_INV_U: Sports Media Involvement; TOT_CONS: Total Sports Media Consumption per Day.) Table 13: Gender Descriptive Statistics Total Sports Media Consumption in Hours per Day Males Females Total N = 222 Mean 4.729 SD 2.501 N = 147 3.877 2.215 N = 369 4.390 2.424 Table 12 illustrates significant results regarding gender differences posed by RQ4. By comparing the path coefficients from the structural model shown earlier in Figure 12, the SmartPLS v.3.2.6 group analysis t-test revealed four separate gender differences. Table 13 reveals that 100 males, on average, consume almost one more hour per day than females (Δ = 0.852 hours per day). Regarding the antecedent motivations, it was found that Escape from Problems (ESC) is significantly higher for males (0.323; p < 0.001) than females (0.115; not significant) whereas Vicarious Achievement (VIC) is significantly higher for females (0.549; p < 0.001) than males (0.276; p < 0.001). For the direct path from team identity to sports media involvement, males have a significantly higher path coefficient (0.286; p < 0.001) than females (0.036; not significant). Finally, for the direct path from sports media involvement to total consumption, males again have a significantly higher path coefficient (0.378; p < 0.001) than females (0.130; not significant). These findings will be inspected further in the discussion section. To supply a real-world interpretation of the standardized coefficients reported throughout this chapter for the analyzed structural model, unstandardized coefficients in the unit of hours were calculated based on the standard deviation of the total sports media consumption per day dependent variable. The reported mean value for time spent consuming sports media per day for the sample of 369 participants was 4.390 hours with a standard deviation of 2.425 hours consumed. These data enable the calculation of predicted values of increased minutes of sports media consumption per day, and hours predicted per year due to the exogenous variables in the model. The results are shown in Table 14. Note that Table 9 revealed three significant pathways predicting increases in total consumption via mediation, plus one direct influence from sports media involvement to total consumption. Those four relationships between predictor constructs and total consumption from Table 9 are shown in Table 14. 101 Table 14: Unstandardized Consumption Coefficients Conversions into Minutes per Day Relationship between constructs SP_INV_U  TOT_CONS TI_U  TOT_CONS VIC  TOT_CONS SOC  TOT_CONS Standardized Total Effects Coefficient Unstandardized Coefficient (Hours) Mins per Day increase in Sports Media Consumption Hours per Year increase in Sports Media Consumption 0.334*** 0.810 48.60 295.85 0.299*** 0.725 43.50 264.81 0.113* 0.274 16.44 100.08 0.203** 0.492 29.52 179.70 *** p < 0.001, ** p < 0.01, * p < 0.05. Table 14 illustrates predicted changes in total sports media consumption in values of minutes consumed per day and hours per year. Forecasting one year’s worth of sports media consumption, the model predicts hundreds of hours of increases consumed per person resulting from changes in the predictor variables. An increase of one unit of sport media involvement (in the 7-unit Likert-type scale) predicts more than a 295-hour consumption increase over the course of year when holding the other variables constant. Second highest is team identity, where a one-unit increase predicts over 264 hours of increased consumption of sports media in a year when holding the other variables constant. Increasing the antecedent motivation of vicarious achievement (VIC) by one unit predicts over 100 hours of increased sports media consumption, and a one-unit increase in social media interaction (SOC) predicts over 179 more hours consumed in a year. The results of the research question analyses are shown in Table 15: 102 Table 15: Research Question Results RQ1: RQ2: RQ3: RQ4: Does Model #1 with Multidimensional Team Identity (Heere & James, 2007b) and Multidimensional Sports Media Involvement (Beaton et al., 2011) explain more variance of Total Sports Media Consumption per day than Model #2 using Unidimensional Team Identity (Wann & Branscombe, 1993) and Unidimensional Sports media involvement? Which first-order facets of Multidimensional Team Identity contribute most to the identity construct? Multidimensional Model (Model #1) not tested – did not exhibit discriminant validity with other latent constructs. Equal Contribute of Influence from all 6 of the first-order multidimensional Team Identity facets. Which first-order facets of Equal Contribute of Multidimensional Sports Media Influence from all 3 of Involvement contribute most to the the first-order involvement construct? multidimensional Sports Media Involvement facets. Are there significant gender differences in ESC – stronger for males the motivational drivers of sports media VIC – stronger for involvement and sports media females consumption? 103 CHAPTER 5 DISCUSSION Interpretation of Structural Model Results This study offers several insights into the daily consumption of sports media and the drivers of increased consumption. It disentangles the pathway of sports media consumption based on the theoretical considerations of Social Identity Theory (Ashforth & Mael, 1989; Tajfel, 1974), Antecedents to Involvement (Funk et al., 2004), and Consumer Involvement in the form of Sports Media Involvement (Beaton et al., 2011; Zaichkowsky, 1985), and illustrates how the relationships work in a linear manner. Prior research has investigated these constructs individually, but never in combination with one another. This study not only incorporates all of the discrete and reliable variables into a significant predictive structural model, but does so using parsimonious unidimensional scales. An R2 of 21.2% of variance explained of total sports media consumption per day is produced by the structural model. Although there is still much variance unexplained, accounting for one-fifth of the time spent on sport media consumption makes this a useful model. The relationships were not due to sampling error, and the process of deciding how to allocate one’s time is a complex process that would require a much more elaborate model to explain a majority of the variance. The overall sample showed a mean of 4.4 hours of sports media consumed per day with a standard deviation of 2.4 hours, indicating that over 68% of the sample consumes between 2.0 and 6.6 hours of sports media per day. This is another important insight into the daily sports media consumption habits of adults in the United States. In conjunction with the explanatory contributions of the path model of sports media consumption, this study pinpoints the construct of sports media involvement for the first time as a unidimensional component. Sports Media Involvement was revealed to consist of feelings 104 combining a centrality of sports media to one’s lifestyle with a symbolic sense of personal relevance that defines how a person sees him or herself. In other words, people who are highly involved with sports media feel that sports media coverage plays such a central role in their lives that it is another way of defining who they are as individuals, much like one’s sense of identity. Sports media involvement can even affect how people feel others see them and their lifestyles. This construct is fueled by feelings that sports media helps one escape from problems (ESC) and allows one to experience vicarious achievement through sports media coverage (VIC). These are the two essential antecedent motivations that predict higher levels of sports media involvement. Engendering those two feelings along with an increased sense of team identity are the three constructs that explain an R2 of 50.0% of sports media involvement, which a moderately high amount of variance explained and is a satisfactory validation of the structural model tested in this study. Furthermore, as the structural model in Figure 12 illustrated, this feeling of increased sports media involvement manifests in a having a large effect on the behavior of watching or following sports media, and in a statistically significant fashion. Sports media involvement is the largest predictor of increased sports media consumption per day in the model, yielding a predicted 48.6 minutes of increased consumption per day, which sums to over 295 additional hours of sports media consumption per year. That is hundreds of additional hours spent watching live sporting events, ESPN SportsCenter and other sports news programming, reading sports magazines and newspapers, following sports online and on one’s Smartphone including social media networks like Twitter and Facebook. All of these mediated forms of communication have some sort of advertising for revenue, which is why increased consumption is desired from every sports media organization. These relationships suggest that sports has become a powerful force 105 in helping people cope with problems in their lives and feel better due to vicarious achievement. It also allows people to feel better about themselves as they escape their problems. The role that sports plays in people’s lives allows media companies to attract readers, viewers and listeners. This study reveals how the process of increased sports media consumption works, and what components make up the overall pathway. The detection of the unidimensional sports media involvement construct is an important discovery for those wishing to study other subsequent behaviors possibly linked to increased involvement in this area. There are most likely other sports-related behavioral outcomes related to increased sports media involvement that can be examined in future studies. The current study reveals that sports media involvement relates positively to a heightened sense of team identity and the antecedent motivations of seeing sports media as a way of escaping from problems and feeling a sense of vicarious achievement provided by the coverage of sports by the media. Based on the significant and positive path coefficients in the structural model shown in Figure 12, it is clear that team identity is the first component of the linear path of increased sports media consumption. Team identity is then followed by the antecedent motivations of Escaping from Problems (ESC), Vicarious Achievement (VIC), and Social Media Interaction (SOC), which then engages Sports Media Involvement and Total Consumption. SOC was discovered to have a moderate direct effect on increased sports media consumption for all individuals, but as an antecedent to involvement, had no significant effect in the model. Sports media involvement was shown to be the strongest predictor of total sports media consumption per day, and in conjunction with the antecedent motivations, fully mediates the relationship between team identity and total consumption. The existence of the antecedent motivations plus sports media involvement increases the variance explained of total consumption by more than 106 double the direct effect of team identity over its direct link shown in Figure 4. The important contribution is that while there are other mediating variables, the total effect of team identity on total sports media consumption per day is still strong and significant at the 0.001 level (see Table 9). This means that team identity level does indeed play an important role in increasing overall sports media consumption, and the structural model of this study shows the intermediate components of that process. The mediation effects via the antecedent motivations (ESC, VIC, & SOC) and sports media involvement are also key findings because they illustrate four discrete constructs that show significant relationships with increased team identity. These variables work together to greatly affect overall sports media consumption. This was shown in the hours per year increases due to these variables (see Table 14). As mentioned earlier in this section, this can be very important from a marketing and sports management perspective. From a communication studies standpoint, the findings about the relationships of these variables show us that once a person feels a sense of belonging to a social group, or in this case, to a favorite sports team, then other processes engage that allow specific motivations to engender even more personal relevance in the form of sports media involvement. Gender Differences in the Consumption of Sports Media Culture is an adaptive tool used by humans where a core set of consistent beliefs and norms are shared by a particular social in-group in order to reduce risk and facilitate group relations (Maslow, 1943; Tajfel, 1974). Sports fan bases have been shown to exhibit similar cultures in the same manner based on a team’s particular in-group (Dionísio et al., 2008; Funk et al., 2004). Commonplace examples include NASCAR racing fan communities who follow their favorite drivers and adhere to commonly held beliefs and behaviors (Billings et al., 2015). 107 Based on the findings of this study regarding gender differences, it can be said that male and female sports media consumers engage with sport media differently. This was one of the central goals of this research study, as it looked to shed light on the psychological reasoning driving sports media involvement and total sports media consumption per day. In this regard, significant gender differences were revealed in both areas. Specifically, it was discovered that the antecedent motivation of Escape from Problems (ESC) was a significantly higher influence on increased sports media involvement for males than females. In fact, the path coefficient was low and nonsignificant between ESC and sports media involvement for females. This indicates that men feel a sense of sanctuary in watching and following sports media that transports them away from the stressful elements of the their lives, which in turn creates more of a feeling of personal relevance dealing with sports media (which subsequently advances male’s increased consumption of sports media each day). Conversely, the antecedent motivation of Vicarious Achievement (VIC) was a significantly higher influence on increased sports media involvement for females than males. The path of VIC to sports media involvement showed a significant (p < 0.001) and moderately high path coefficient (0.276) for males as well, but the path coefficient of 0.549; p < 0.001 (see Table 12 for reference) was the largest of the entire model for any variable relationship for females only. This indicates that women are driven to feel a sense of more personal relevance about sports media because of the antecedent feelings of vicariously achieving the winning sports achievements watched or followed on sports media. Each of the items in the 3-item scale concerned experiencing the victories of athletes and teams in an indirect and surrogate manner, so this tells us that while it is essential for all individuals, it is the central motivation fostering sports media involvement for females. 108 Building off the importance of the vicarious achievement motivation for females, it could be said that this construct of VIC fully mediates the relationship between team identity and sports media involvement for women but not men. This is seen in the almost nonexistent (0.036) path coefficient from team identity to sports media involvement for females, which was unsurprisingly also nonsignificant. For males, the direct path from team identity to sports media involvement is both larger than females in terms of the path coefficient (0.286) and significant at the 0.001 level, making it a third gender difference discovered by this study regarding sports media. The final significant gender difference was seen in the direct path between sports media involvement and total sports media consumption per day. For males, the relationship was large with a path coefficient of 0.378 and significant at p < 0.001, but for females, the coefficient was much lower at 0.130 and nonsignificant. The interpretation is that the model of sports media consumption tested in this study shows that males’ sports media consumption is positively affected by increases in team identity, escape from problems, vicarious achievement, social media interaction, and sports media involvement. For female sports media consumers, increased feelings of team identity are fully mediated by strong feelings of vicarious achievement provided by sports media, which then engenders an increase in the feeling of sports media involvement. This feeling does not manifest in significant predicted increases in total consumption for female sports media consumers. Future research is necessary to determine the subsequent behavioral ramifications for this population of sports media consumers due to an increase in sports media involvement. Women are sports consumers and do watch and follow sports media, but could be consuming something alternative to what was measured in this study. 109 Males and females shared similarities in the way increases in team identity level predicted increases in the three antecedent motivations (ESC, VIC, and SOC), as well as the total effect of increased team identity on increased time spent consuming sports media. Additionally, the motivational antecedent of vicarious achievement significantly correlated with increases in sports media involvement for both men and women, although this path was significantly higher for females. Finally, the antecedent motivation of social media interaction predicted increases in time spent consuming sports media for both males and females. These matching relationships can be useful for marketing messages aimed at all US adult sports media consumers, and allows for a psychological understanding of this group of individuals. Overall, several important significant gender differences regarding sports media involvement and sports media consumption were revealed by the structural model tested in this study. It is important to not only isolate what specific motivations relate to all sports fans, but to understand the mechanisms involved sports media male and female sports fans. This provides increased knowledge of how consumers engage with sports media, enabling social science research to better study the mechanisms by which individuals process being part of a sports ingroup, and how this affects levels of involvement and the behavior of sports media consumption. Note that other demographic data was collected in this study, such as age, education, income, race, and marital status, but there were no other significant differences in the structural model found between groups of sports fans based on these categories. 110 Managerial and Marketing Implications The information provided by this research project can assist sports marketing managers who choose to follow the implications discussed in this chapter. Solutions to how an organization should best reach its audience – and how this will affect sports media consumption – can be very helpful and potentially profitable based on the findings of this study. Marketing managers could use the findings to craft specific advertising messages to all sports consumers that foster the feeling of team identity, motivations like escaping from problems, vicarious achievement, and social media interaction. For males, it would be wise to center messages on the motivation to escape from problems, as this was shown to be significantly higher for men than women. For females, the recommendation would be to craft messages that exemplify the feelings of vicarious achievement experienced from watching or following sports media. Both males and females showed significant correlations between increases in the vicarious achievement motivation and sports media involvement, so it would be prudent for marketing managers to consider this for possible mediated messages disseminated to sports media consumers to encourage additional consumption. Following the linear steps of the process of fostering increased sports media consumption for males was shown to predict hundreds of hours of increased sports media consumption per year. This would be very beneficial from a marketing perspective, especially when one considers the millions of individuals consuming sports media on a daily basis throughout the year. Furthermore, this study has shown how the mechanisms of engendering increased sports media involvement, which surely has other marketing ramifications other than watching or following sports media (such as possibly purchasing merchandise or attending live sporting events). These 111 reasons illustrate the significance of the research model in the area of sports management and marketing. The results of this study provide a detailed understanding of what types of motivations are significant in predicting a change in sports media consumption, while considering gender and revealing significant differences between male and female sports media consumers. This information can be used to craft precise advertising messages geared towards specific sports audiences in order to nurture potentially profitable outcomes like increased sports media consumption. The more time an organization can get an audience to spend watching and following their media means more money in the pockets of sports media companies thanks to the positive relationship between advertising rates and the increased consumption of media (Pegoraro, Ayer, & O'Reilly, 2010). Social Identity Theory and Involvement Implications Overall, the results of this study are consistent with Social Identity Theory and the theory of Consumer Involvement, as well as the relationship between the two concepts. This can be seen in the linear way the constructs significantly correlate and predict changes in the outcome variables. Increased team identity increases each of the antecedent motivations, and two of these antecedents increase sports media involvement. Those psychological components were statistically shown to significantly correlate with increases in time spent watching or following sports media. In its simplest terms, sports team identity and the personal relevance of being involved with sports media are the key ingredients affecting increases in the behavior of consuming sports media. As mentioned earlier, this study adds to the knowledge of sports media engagement from a communication research standpoint by revealing the social and psychological impact of sports 112 media consumption. Positive aspects of identifying with a sports team have been found to include enhanced levels of self-esteem, reduced feelings of depression, and a lowered sense of alienation from the rest of society (Branscombe & Wann, 1991). These outcomes could be directly affected by the findings that escaping from problems, vicarious achievement, and social media interaction play important roles in increasing a sense of being involved with sports media (in a way that is central to one’s life that also has symbolic value) and increasing the amount of time spent consuming media. It is possible that testing for a feedback loop in a non-recursive model would reveal that increased time spent consuming correlates with increases in one’s sense of team identity. Additionally regarding social identity, having a personal connection with a sports team had been shown to offer an enhanced sense of belongingness to the community at large and the outlook that sports fans feel they are involved in something greater than themselves, especially for youngsters participating in sports (Eime et al., 2013; Theokas, 2009; Zaff et al., 2003; Zarrett et al., 2009). By understanding the results shown in this research study, this information can help scholars recognize the mechanisms that result from increased team identity and the outcomes of such feelings. Thus, this study helps to explain some of the psychological outcomes that drive feelings of belongingness and other social and psychological impacts of consuming sports media. It has been said that being part of an online or social media community is similar to being a physical member a community, much like that of a team or being a sports spectator in person (Gibbons & Dixon, 2010; Mangold & Faulds, 2009; McCarthy, 2012). This makes the finding of increased social media interaction as a consequence of higher team identity an important aspect of this study. 113 Increases in team identity influences increases in both the motivation to feel that sports media can provide an escape from problems in one's daily life as well as sense that sports media is a way of experiencing vicarious achievement. For males, this sense of escaping from problems is a significant influence on increased sports media involvement, as is the feeling of vicarious achievement. For females, only vicarious achievement significantly influences an increase in sports media involvement while the escape from problems motivation shows very little effect on the construct. Therefore, the sociological pathway from increased team identity level to increased sports media involvement is different for males and females due to the psychological motivations driving the increases in involvement. This is an interesting finding to determine the social consequences of increased sports team identity based on gender. Additionally, the increase in sports media involvement was shown to have a significant effect on increased behavior in the form of consuming sports media for males alone. For females, the increased involvement likely has other downstream effects besides increases in sports media consumption. This points to a need to deduce what specifically predicts increases in sports media consumption behavior for females and can direct future study in this area. It is possible that factors such as the presence of other family members or those in the same social in-group of the sports team affects increased consumption of sports media for females and males alike. Finally, this study adds to social identity theory literature by illustrating the consequential outcomes of an increased sense of sports team identity. This not only manifests in the significant behavioral outcome of increased time spent watching and following sports media, it significantly engenders several social and psychological outcomes. These social outcomes include seeing sports media as a way of escaping from problems, vicariously achieving success, and a desire to interact with other sports fans via social media platforms like Twitter and Facebook. This project 114 contributes to involvement literature by providing a unidimensional scale of sports media involvement that is both reliable and valid. This newly discovered scale can be used in future research projects to determine additional consequences of increased sports media involvement other than increased sports media consumption. These outcomes can be both psychological and behavioral. Lastly, this study reveals differences in the motivational antecedents driving increased sports media involvement for males and females. This can be useful for future research regarding how team identity and sports media involvement affect men and women differently, which will help deepen our understanding of sports fans and spectators, as well as sports media consumers. It would be interesting to test to see if sports team identification is incorporated into a type of hierarchical identity structure for individuals when compared to other similar psychological connections such as religious and ethnic identity. A discovery that individuals put more weight on specific in-groups over others, and what would cause such an ordinal categorization arrangements, could add to the depth of knowledge regarding the inspection of social identity. To conclude, humans have a need to answer questions about their own identities so that they can establish who they are and what their lives represent. Social Identity Theory aids in the understanding of such uncertainty reduction practices, and involvement reflects humans’ need to engage with things that are personally relevant to who they are as individuals. This study has added to the knowledge of how the mechanisms of identity and involvement relate to sports media consumers in the digital age of the 21st Century. 115 Limitations and Future Research One of the main limitations of this study was in the measurement of the dependent variable of Total Sports Media Consumption per Day. This was a summed variable consisting of five self-reported values of time spent consuming sports media each day on the following types of media: television, newspapers or magazines, the Internet, radio or podcasts, and on social media. The most deficient aspect of this variable with regards to media consumption is that it does not ask specifically what media consumers are paying attention to when they do consume. It could be very possible this could have affected the gender differences as men and women could very well be doing different things behaviorally when consuming that would affect how marketing messages could affect them. More importantly, it prevents a thorough sociological examination of what it means to consume sports media each day. Due to this issue, future investigation into sports media consumption should focus not only on time spent consuming, but what specifically is being consumed within each of the types of media that sports fans follow or watch. This increased focus will allow for more close scrutiny of the mechanisms of sports media consumption and the motivations and predictor variables driving changes in time spent consuming. In addition, testing the structural model seen in Figure 12 with a new set of data using covariance-based SEM would allow for the investigation of any feedback loops (a non-recursive model) in order to inspect whether heightened sports media involvement and increased time spent consuming sports media affects increases in team identity level. This would shed light on any social and psychological outcomes in this study’s predictor variables from increased behaviors consuming sports media. Further research that expands the implications of Social Identity Theory and Consumer Involvement could possibly incorporate other exogenous predictor variables into the structural 116 model to determine if more of the variance can be explained, raising R2 levels of the endogenous constructs. Such variables could include scales of fan engagement (Hollebeek, Glynn, & Brodie, 2014; Sashi, 2012), absorption (Schaufeli, Salanova, González-Romá, & Bakker, 2002), or interactivity (Liu & Shrum, 2002). Related to the issue of the dependent variable of Total Sports Media Consumption per Day, with an R2 of 21.2% seen in the final model in Figure 12, there is a majority of that variance that was not explained. Replicating this study with other variables such as engagement, absorption, and interactivity incorporated into the model could possibly increase the amount of variance explained, but it would seem that a more complex measure of sports media consumption (direct observation rather than self-reporting, perhaps) could provide a clearer understanding of how the exogenous variables in the research model affect sports media consumption. Furthermore, a future study that validates the unidimensional sports media involvement scale will substantiate its reliability and illustrate its potential usefulness in sports media studies dealing with consumer behavior. Likewise, since the sample was taken from a cross-sectional representative collection of United States adults, there are limitations to the overall generalizability of the results to the entire population. The screener question required participants to indicate if they were regular sports media consumers, so this study can be generalized to those individuals who regularly consume sports media on TV, newspapers or magazine, the radio or podcasts, on the Internet, and via social media. This sample did provide a valid look at the psychological motivations of adult sports fans in the United States. Future research should investigate the possibility that the findings in this study apply to other behavioral outcomes such as attending sporting events or purchasing team-related 117 merchandise. Sports media involvement is a strong predictor of increased time spent consuming sports media for males but did not show a significant correlation in this respect for females. Testing for additional consequences of increased sports media involvement will enable scholars to understand the ramifications of this feeling as it pertains to other social and psychological outcomes. Additionally, in lieu of the finding that females did not have a significant relationship between the motivation of escaping from problems and sports media involvement (with males being significantly higher regarding the motivation of escaping from problems while also being high in the motivation of vicarious achievement), the question arises as to what other factors besides team identity and vicarious achievement are at play while women are consuming sports. The bonding with friends motivation (BON) did not result in a discriminant factor in the measurement model, but it could be possible that females are consuming sports in the presence of friends and family members more than males. This supposition is a result of the male-only significant escape from problems motivation that could possibly suggest solo consumption of sports media. Future research should investigate whether males and females report different consumption habits regarding the presence of friends and/or family members during group sports media consumption. While the social media interaction showed significant correlations as a link predicting increased time spent consuming sports media, perhaps the presence of others in a physical form might be a possible explanation for increased sports media consumption and higher levels of sports media involvement for females. Future examination in this area would enrich the findings of the current project. In conclusion, this study contributes to the investigation of sports media consumers and their relationship with sports media. It provides a statistical verification of the connections 118 between team identity level, antecedent motivations, sports media involvement, and changes in time spent consuming sports media each day. In addition, this project provides details on gender differences for adult sports media consumers and the motivational differences in this process between men and women. The discovery of the unidimensional sports media involvement scale is an important source for future research related to sports media, and can help the investigation of sports media consumption from both a communication and sports marketing and management perspective. Sports media is a billion-dollar industry that will surely exist for the foreseeable future. This study has provided an inside look at how individuals relate to the consumption of sports media in the 21st Century. 119 APPENDIX 120 APPENDIX Survey Instrument Sports Media Involvement via Team Identity & Antecedent Motivations for the Prediction of Total Daily Sports Media Consumption Section 1: 5 Sports Media Consumption Amount Items total: Note to participant: For questions #1-5, please think about how much time you spend per day watching or following sports on a the listed media platform for that question. Note to participant: “Sports media” includes live or recorded sporting events, sports news or highlights, sports talk shows or podcasts, social media for sports (via Twitter & Facebook), and/or sports programming such as sports documentaries. 1) In an average day, I spend this amount of time watching or following sports on Television…        Zero minutes per day Less than 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours Etc. (until 11-12 hours) 2) In an average day, I spend this amount of time reading about sports in newspapers or magazines…        Zero minutes per day Less than 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours Etc. (until 11-12 hours) 121 3) In an average day, I spend this amount of time watching or following sports on my laptop, Smartphone or portable device…        Zero minutes per day Less than 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours Etc. (until 11-12 hours) 4) In an average day, I spend this amount of time listening to sports talk radio and/or sports podcasts (on the actual radio or using a Smartphone app)…        Zero minutes per day Less than 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours Etc. (until 11-12 hours) 5) In an average day, I spend this amount of time following and/or discussing sports on social media (e.g. Twitter, Facebook, Snapchat, etc.)…       Zero minutes per day Less than 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours  Etc. (until 11-12 hours) Section 2: 4 Unidimensional Team Identity items total: EACH TEAM IDENTITY ITEM IS MEASURED WITH A 7-POINT LIKERT SCALE. Note to participant: For the following 4 questions, please answer regarding your feelings about yourself as a sports fan. 122 Note to participant: “Sports media” includes live or recorded sporting events, sports news or highlights, sports talk shows or podcasts, social media for sports (via Twitter, Facebook & Snapchat), and/or sports programming such as sports documentaries. 4 items based on a scale of Unidimensional Team Identity: 1) How important to you is it that your favorite sports team wins? 2) How strongly do you see yourself as a fan of your favorite sports team? 3) How strongly do you see yourself as a fan of your favorite sports team? 4) How important is being a fan of your favorite sports team to you? Table 16: Unidimensional Team Identity Items Construct Unidimensional Team Identity (TI_U) Item 4 TI_U items Question Feelings of Unidimensional Team Identity (1=Strongly Disagree; 7= Strongly Agree) TI_U1 How important to you is it that your favorite sports team wins? How strongly do you see yourself as a fan of your favorite sports team? How strongly do your friends see you as a fan of your favorite sports team? How important is being a fan of your favorite sports team to you? TI_U2 TI_U3 TI_U4 Source (Wann & Branscombe, 1993) Section 3: 18 Multidimensional Team Identity items total: EACH TEAM IDENTITY ITEM IS MEASURED WITH A 7-POINT LIKERT SCALE. Note to participant: For the following 18 questions, please answer regarding your feelings about yourself as a sports fan. Note to participant: “Sports media” includes live or recorded sporting events, sports news or highlights, sports talk shows or podcasts, social media for sports (via Twitter, Facebook & Snapchat), and/or sports programming such as sports documentaries. 6 first-order facets of Multidimensional Team Identity x 3 items each: 123 Table 17: Multidimensional Team Identity Items Construct Multidimensional Team Identity (TI_M) Item 18 TI_M items Question Feelings regarding Multidimensional Team Identity (1=Strongly Disagree; 7= Strongly Agree) Source Public Evaluation TI_M_PbE1 Overall, my favorite team is viewed positively by other people. In general, other people respect my favorite team. People hold favorable opinions about my favorite team. I feel good about cheering for my favorite team. (Heere et al., 2011) via (Luhtanen & Crocker, 1992) TI_M_PbE2 TI_M_PbE3 Private Evaluation TI_M_PvE1 TI_M_PvE2 TI_M_PvE3 Sense of Interdependence with the Group TI_M_IntD1 TI_M_IntD2 TI_M_IntD3 Interconnection with the Group TI_M_IntC1 TI_M_IntC2 TI_M_IntC3 Behavioral Involvement TI_M_B1 TI_M_B2 TI_M_B3 In general, I’m glad to be a fan of my favorite team. I am proud to think of myself as a fan of my favorite team. The performance of my favorite team influences what happens in my life. Changes that affect my favorite team have an impact on my life. What happens to my favorite team has an impact on my life. When someone criticizes my favorite team, it feels like a personal insult. Being associated with my favorite team is an important part of my self-image. When someone says something positive about my favorite team, it feels like a personal compliment to me. I participate in activities supporting my favorite team. I am actively involved with any activities related to my favorite team. I participate in activities with other fans of my favorite team. 124 (Heere et al., 2011) via (Luhtanen & Crocker, 1992) (Heere et al., 2011) via (Gurin & Townsend, 1986) (Heere et al., 2011) via (Mael & Tetrick, 1992) (Heere et al., 2011) via (Phinney, 1992) Table 17 (cont’d) Cognitive Awareness TI_M_CA1 TI_M_CA2 TI_M_CA3 I am aware of the traditions of my favorite team. I am aware of the specific details regarding my favorite team (e.g.: history, win/loss record, current and past players, coaches, etc.). I am familiar with the past successes and failures of my favorite team. (Heere et al., 2011) Section 4: 36 Sports Media Consumption Motivation items total: EACH SPORTS MEDIA CONSUMPTION MOTIVATION ITEM IS MEASURED WITH A 7POINT LIKERT SCALE. Note to participant: For the following 36 questions, please answer regarding your sports media consumption motivations. Note to participant: “Sports media” includes live or recorded sporting events, sports news or highlights, sports talk shows or podcasts, social media for sports (via Twitter, Facebook & Snapchat), and/or sports programming such as sports documentaries. 12 Motivational Factors x 3 items each: Table 18: Antecedent Motivation Items Construct Motivations to Consume Sports Media Item 36 Motivation items PLA1 (Interest in Players) PLA2 (Interest in Players) PLA3 (Interest in Players) BON1 (Bonding with Friends) BON2 (Bonding with Friends) Question Feelings regarding Motivations to Consume Sports Media (1=Strongly Disagree; 7= Strongly Agree) I cheer for one or more athletes that I consider my favorite. I will follow my favorite player no matter which team he or she plays on. I like my favorite player more than the team he or she plays for. I feel that sports media help me bond with my friends. Sports media allow for a shared experience for me with my friends. 125 Source (Funk, Ridinger, Moorman, 2004) & Table 18 (cont’d) BON3 (Bonding with Friends) DRA1 (Drama - Sports Media Updates) DRA2 (Drama - Sports Media Updates) DRA3 (Drama - Sports Media Updates) COM1 (Community Connectedness) Sports are more enjoyable when my friends and I are watching together. I enjoy getting live sports updates through sports media. COM2 (Community Connectedness) COM3 (Community Connectedness) ROL1 (Role Models) The various types of sports media help me to feel part of the sports fan community. Sports media are important in my life because they connect me to the sports fan community. I feel that collegiate and professional athletes are good role models for me and others. Successes by other athletes inspires me to become more successful on my own. Athletes are people that inspire me to be a better person. Watching and following sports media is exciting to me. ROL2 (Role Models) ROL3 (Role Models) EXC1 (Exciting Coverage of my Team) I like checking sports media for updates about live games. I like seeing sports media alerts about breaking news. Sports media make me feel connected to the sports fan community. EXC2 (Exciting Coverage of my Team) Sports media help to make sports more exciting. EXC3 (Exciting Coverage of my Team) Sports media provide a constant source of excitement in my life. KNO1 (Knowledge of Sports) KNO2 (Knowledge of Sports) Sports media help to expand my knowledge about sports in general. I like when sports media provide me with statistical information about players, teams, and game stats. The information provided by sports media make sports more enjoyable to me. Sports media that focuses on winners and champions makes me feel like a winner too. Sports media makes me feel a sense of personal accomplishment when I experience sports media coverage of winners and champions. KNO3 (Knowledge of Sports) VIC1 (Vicarious Achievement) VIC2 (Vicarious Achievement) 126 Table 18 (cont’d) VIC3 (Vicarious Achievement) SOC1 (Social Media Interaction) SOC2 (Social Media Interaction) SOC3 (Social Media Interaction) ESC1 (Escape from Problems) ESC2 (Escape from Problems) ESC3 (Escape from Problems) FAM1 (Family Atmosphere) FAM2 (Family Atmosphere) FAM3 (Family Atmosphere) TEM1 (Team Interest – Behind the Scenes) TEM2 (Team Interest – Behind the Scenes) TEM3 (Team Interest – Behind the Scenes) Sports media coverage of success in sports is the main reason why I watch and follow sports. I like when sports media allow me to interact with other fans via social media. I like communicating with other sports fans in an open public forum via social media. I like when sports media allow me to voice my opinion about players and teams. Sports media distract me from the bad or annoying things in my everyday life. Sports media are part of my daily or weekly routine because they allow me to escape my problems. I can get away from the problems in my life by watching sports or following sports media. Sports media creates a family-type atmosphere when I watch and/or follow sports. I feel that other sports media personalities are like family members to me. Watching and/or following my favorite sports media makes me feel like part of the sports media family. Sports media provide a behind-thescenes view of teams and players to me. I like when sports media show me the inner workings and lives of teams and players. I enjoy how sports media give me an all-access look at my favorite players and teams. Section 5: 9 items based on a scale of Multidimensional Sports Media Involvement: EACH SPORTS MEDIA INVOLVEMENT ITEM IS MEASURED WITH A 7-POINT LIKERT SCALE. Note to participant: For the following 9 questions, please answer regarding your thoughts & feelings about sports media. 127 Note to participant: “Sports media” includes live or recorded sporting events, sports news or highlights, sports talk shows or podcasts, social media for sports (via Twitter, Facebook & Snapchat), and/or sports programming such as sports documentaries. 3 first-order facets of Multidimensional Sports Media Involvement x 3 items each: Table 19: Sports Media Involvement Items Construct Sports Media Involvement Facets Item 9 SP_INV Items Hedonic Value SP_INV_HV1 SP_INV_HV2 SP_INV_HV3 Centrality to Lifestyle SP_INV_C1 SP_INV_C2 SP_INV_C3 Symbolic Value SP_INV_SV1 SP_INV_SV2 SP_INV_SV3 Question Feelings of Sports Media Involvement (1=Strongly Disagree; 7= Strongly Agree) It is fun to watch/follow sports media. Watching/following sports media are some of the most satisfying things in my life. I really enjoy watching/following sports media. A lot of my life is organized around watching and/or following sports media. Watching and/or following sports media plays a central role in my life. I enjoy discussing what I read/see/hear on sports media with others. Watching and/or following sports media says a lot about who I am. Watching and/or following sports media tells others about me as a person. My habits of watching/following sports media gives a glimpse of the type of person I am. 128 Source (Beaton et. al, 2011) Section 6: 5 Demographic Items: 1) What is your gender?  Male  Female 2) How old are you?  18 - 29  30 – 49  50 – 59  60+ years old 3) What is the highest level of education completed?  Did not graduate high school  High School Diploma (or GED)  Associate’s Degree  Bachelor’s Degree  Master’s Degree  PhD, MD, JD, DO, EDD 4) What is your race?  Asian  Black or African-American  Native American  Pacific Islander  White or Caucasian  Hispanic or Latino 5) What is your marital status?  Single, never married  Currently married or in a domestic partnership  Widowed  Divorced  Separated END OF SURVEY 129 BIBLIOGRAPHY 130 BIBLIOGRAPHY Adweek.com. (2014). NFL Scores a Touchdown With Female Fans. Retrieved January 28, 2015, from http://www.adweek.com/news/press/nfl-scores-touchdown-female-fans159674 Ahlers, D. (2006). News Consumption and the New Electronic Media. The Harvard International Journal of Press/Politics, 11(1), 29-52. Althaus, S. L., & Tewksbury, D. (2000). Patterns of Internet and Traditional News Media Use in a Networked Community. Political Communication, 17(1), 21-45. Alvermann, D. E., & Hagood, M. C. (2000). Fandom and Critical Media Literacy. Journal of Adolescent & Adult Literacy, 43(5), 436-446. Antil, J. H. (1984). Conceptualization and Operationalization of Involvement. Advances in Consumer Research, 11(1), 203-209. Armstrong, K. L. (2002a). An Examination of the Social Psychology of Blacks' Consumption to Sport. Journal of Sport Management, 16(4), 267-288. Armstrong, K. L. (2002b). Race and Sport Consumption Motivations: A preliminary investigation of a Black consumers' sport motivation scale. Journal of Sport Behavior, 25(4), 309-330. Arnett, D. B., & Laverie, D. A. (2000, 2000 September-October). Fan Characteristics and Sporting Event Attendance: Examining variance in attendance. International Journal of Sports Marketing & Sponsorship, 2, 219-244. Ashforth, B. E., & Mael, F. (1989). Social Identity Theory and the Organization. The Academy of Management Review, 14(1), 20-39. Ashmore, R. D., Deaux, K., & McLaughlin-Volpe, T. (2004). An Organizing Framework for Collective Identity: Articulation and significance of multidimensionality. Psychological Bulletin, 130(1), 80-114. Barki, H., & Hartwick, J. (1994). Measuring User Participation, User Involvement, and User Attitude. MIS Quarterly, 18(1), 59-82. Baron, R. M., & Kenny, D. A. (1986). The Moderator–Mediator Variable Distinction in Social Psychological Research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182. 131 Bauer, H. H., Stokburger-Sauer, N. E., & Exler, S. (2008). Brand Image and Fan Loyalty in Professional Team Sport: A refined model and empirical assessment. Journal of Sport Management, 22(2), 205-226. Beard, J. G., & Ragheb, M. G. (1983). Measuring Leisure Motivation. Journal of Leisure Research, 15(3), 219. Beaton, A. A., & Funk, D. C. (2008). An Evaluation of Theoretical Frameworks for Studying Physically Active Leisure. Leisure Sciences, 30(1), 53-70. Beaton, A. A., Funk, D. C., Ridinger, L. L., & Jordan, J. (2011). Sport Involvement: A conceptual and empirical analysis. Sport Management Review, 14(2), 126-140. Bell, B. T., & Dittmar, H. (2011). Does Media Type Matter? The role of identification in adolescent girls’ media consumption and the impact of different thin-ideal media on body image. Sex Roles, 65(7), 478-490. Bennett, G., Sagas, M., & Dees, W. (2006). Media Preferences of Action Sports Consumers: Differences between Generation X and Y. Sport Marketing Quarterly, 15(1), 40-49. Bergami, M., & Bagozzi, R. P. (2000). Self-categorization, Affective Commitment and Group Self-esteem as Distinct Aspects of Social Identity in the Organization. British Journal of Social Psychology, 39(4), 555-577. Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating Online Labor Markets for Experimental Research: Amazon.com's Mechanical Turk. Political Analysis, 20(3), 351368. Billings, A. C., Butterworth, M. L., & Truman, P. D. (2015). Communication and Sport: Surveying the field (2nd ed.). Los Angeles, CA: Sage. Bloch, P. H., & Bruce, G. D. (1984). Product Involvement as Leisure Behavior. Advances in Consumer Research, 11(1), 197-202. Boninger, D. S., Krosnick, J. A., & Berent, M. K. (1995). Origins of Attitude Importance: Selfinterest, social identification, and value relevance. Journal of Personality and Social Psychology, 68(1), 61-80. Branscombe, N. R., & Wann, D. L. (1991). The Positive Social and Self Concept Consequences of Sports Team Identification. Journal of Sport & Social Issues, 15(2), 115-127. Branscombe, N. R., Wann, D. L., Noel, J. G., & Coleman, J. (1993). In-Group or Out-Group Extremity: Importance of the threatened social identity. Personality and Social Psychology Bulletin, 19(4), 381-388. 132 Brasel, S. A., & Gips, J. (2011). Media Multitasking Behavior: Concurrent television and computer usage. Cyberpsychology, Behavior, and Social Networking, 14(9), 527-534. Brewer, M. B., & Gardner, W. (1996). Who is this "We"? Levels of collective identity and self representations. Journal of Personality and Social Psychology, 71(1), 83-93. Brown, J. A. (1997). Comic Book Fandom and Cultural Capital. The Journal of Popular Culture, 30(4), 13-31. Brown, M. T., Zuefle, D. M., & Batista, P. J. (2007). Will the Real 12th Man Please Stand Up? Texas A&M and Seattle Seahawks settle dispute over right to identify fans as '12th Man'. Sport Marketing Quarterly, 16(2), 115-117. Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6(1), 3-5. Bush, V. D., Bush, A. J., Clark, P., & Bush, R. P. (2005). Girl Power and Word-of-Mouth Behavior in the Flourishing Sports Market. Journal of Consumer Marketing, 22(5), 257264. Bureau of Labor Statistics; United States Department of Labor (2014). Leisure and Sports on Weekdays and Weekend Days in 2013. Retrieved May 18, 2016, from http://www.bls.gov/tus/current/leisure.htm Bureau of Labor Statistics; United States Department of Labor (2015). Charts by Topic: Leisure and sports activities (leisure time on an average day). Retrieved May 18, 2016, from http://www.bls.gov/tus/charts/leisure.htm Carroll, B., & Alexandris, K. (1997). Perception of Constraints in Strength of Motivation: Their relationship to recreational sport participation in Greece. Journal of Leisure Research, 29(3), 279. Casler, K., Bickel, L., & Hackett, E. (2013). Separate but Equal? A comparison of participants and data gathered via Amazon’s MTurk, social media, and face-to-face behavioral testing. Computers in Human Behavior, 29(6), 2156-2160. Casper, J. M., Gray, D. P., & Babkes Stellino, M. (2007). A Sport Commitment Model Perspective on Adult Tennis Players’ Participation Frequency and Purchase Intention. Sport Management Review, 10(3), 253-278. Casper, J. M., & Stellino, M. B. (2008). Demographic Predictors of Recreational Tennis Participants' Sport Commitment. Journal of Park & Recreation Administration, 26(3), 95-115. 133 Chan-Olmsted, S., Rim, H., & Zerba, A. (2013). Mobile News Adoption among Young Adults: Examining the roles of perceptions, news consumption, and media usage. Journalism & Mass Communication Quarterly, 90(1), 126-147. Chan, J. K., & Leung, L. (2005). Lifestyles, Reliance on Traditional News Media and Online News Adoption. New Media & Society, 7(3), 357-382. Cheung, G. W., & Rensvold, R. B. (2002). Evaluating Goodness-of-Fit Indexes for Testing Measurement Invariance. Structural Equation Modeling: A Multidisciplinary Journal, 9(2), 233-255. Chin, W. W. (1998). Issues and Opinion on Structural Equation Modeling, Editorial, MIS Quarterly, 1-11. Chin, W. W. (2010). How to Write Up and Report PLS Analyses. In Vinzi, V. E., Chin, W. W., Henseler, J., & Wang, H. (Eds.), Handbook of Partial Least Squares: Concepts, methods and applications (655-690). New York, NY: Springer. Chyi, H. I., & Lee, A. M. (2013). Online News Consumption. Digital Journalism, 1(2), 194-211. Cialdini, R. B., Borden, R. J., Thorne, A., Walker, M. R., Freeman, S., & Sloan, L. R. (1976). Basking in Reflected Glory: Three (football) field studies. Journal of Personality and Social Psychology, 34(3), 366-375. Costley, C. L. (1988). Meta-Analysis of Involvement Research. Advances in Consumer Research, 15(1), 554-562. Creedon, P. J. (1998). Women, Sport, and Media Institutions: Issues in sports journalism and marketing. In Wenner, L. A. (Ed.), MediaSport (pp. 88-99). New York, NY: Routledge. Dietz-Uler, B., Harrick, E., End, C., & Jacquemotte, L. (2000). Sex Differences in Sport Fan Behavior and Reasons for Being a Sport Fan. Journal of Sport Behavior, 23(3), 219-231. Dimmock, J. A., Grove, J. R., & Eklund, R. C. (2005). Reconceptualizing Team Identification: New dimensions and their relationship to intergroup bias. Group Dynamics: Theory, Research, and Practice, 9(2), 75-86. Dimmock, J. A., & Gucciardi, D. F. (2008). The Utility of Modern Theories of Intergroup Bias for Research on Antecedents to Team Identification. Psychology of Sport and Exercise, 9(3), 284-300. Dionísio, P., Leal, C., & Moutinho, L. (2008). Fandom Affiliation and Tribal Behaviour: A sports marketing application. Qualitative Market Research: An International Journal, 11(1), 17-39. 134 Dix, S., Phau, I., & Pougnet, S. (2010). “Bend it like Beckham”: The influence of sports celebrities on young adult consumers. Young Consumers, 11(1), 36-46. Dunnett, C. W. (1964). New Tables for Multiple Comparisons with a Control. Biometrics, 20(3), 482-491. Dupont, W. D., & Plummer Jr., W. D. (1998). Power and Sample Size Calculations for Studies Involving Linear Regression. Controlled Clinical Trials, 19(6), 589-601. Dutta-Bergman, M. J. (2004). Complementarity in Consumption of News Types Across Traditional and New Media. Journal of Broadcasting & Electronic Media, 48(1), 41-60. Eagly, A. H., & Chaiken, S. (2007). The Advantages of an Inclusive Definition of Attitude. Social Cognition, 25(5), 582-602. Eckel, C. C., & Grossman, P. J. (2005). Managing Diversity by Creating Team Identity. Journal of Economic Behavior & Organization, 58(3), 371-392. Edwards, J. R. (2001). Multidimensional Constructs in Organizational Behavior Research: An integrative analytical framework. Organizational Research Methods, 4(2), 144-192. Efron, B. (1987). Better Bootstrap Confidence Intervals. Journal of the American Statistical Association, 82(397), 171-185. Eime, R. M., Young, J. A., Harvey, J. T., Charity, M. J., & Payne, W. R. (2013). A Systematic Review of the Psychological and Social Benefits of Participation in Sport for Children and Adolescents: Informing development of a conceptual model of health through sport. The International Journal of Behavioral Nutrition and Physical Activity, 10(98), 1-21. Fisher, R. J., & Wakefield, K. (1998). Factors Leading to Group Identification: A field study of winners and losers. Psychology & Marketing, 15(1), 23-40. Foehr, U. G. (2006). Media Multitasking among American Youth: Prevalence, predictors and pairings. Henry J. Kaiser Family Foundation, 1-39. Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the Use of Exploratory Factor Analysis in Psychological Research. Psychological Methods, 4(3), 272-299. Forbes.com. (2014). Untapped Resource: The Power of Women in the Sports Marketplace. Retrieved January 11, 2017, from http://www.forbes.com/sites/darrenheitner/2015/03/17/the-march-madness-advertisingbusiness-is-booming/ 135 Forbes.com. (2015). The World's Most Valuable Brands 2015: Behind the numbers. Retrieved April 21, 2016, from http://www.forbes.com/sites/kurtbadenhausen/2015/05/13/the-worlds-most-valuablebrands-2015-behind-the-numbers/#503cd61327c5 Fullerton, S., & Merz, G. R. (2008). The Four Domains of Sports Marketing: A conceptual framework. Sport Marketing Quarterly, 17(2), 90-108. Funk, D. C., & James, J. D. (2001). The Psychological Continuum Model: A conceptual framework for understanding an individual's psychological connection to sport. Sport Management Review, 4(2), 119-150. Funk, D. C., & James, J. D. (2006). Consumer Loyalty: The meaning of attachment in the development of sport team allegiance. Journal of Sport Management, 20(2), 189-217. Funk, D. C., Mahony, D. F., Nakazawa, M., & Hirakawa, S. (2001). Development of the Sport Interest Inventory (SII): Implications for measuring unique consumer motives at team sporting events. International Journal of Sports Marketing & Sponsorship, 3, 291-316. Funk, D. C., Mahony, D. F., & Ridinger, L. L. (2002). Characterizing Consumer Motivation as Individual Difference Factors: Augmenting the sport interest inventory (SII) to explain level of spectator support. Sport Marketing Quarterly, 11(1), 33-43. Funk, D. C., Ridinger, L. L., & Moorman, A. M. (2004). Exploring Origins of Involvement: Understanding the relationship between consumer motives and involvement with professional sport teams. Leisure Sciences, 26(1), 35-61. Gantz, W., & Wenner, L. A. (1991). Men, Women, and Sports: Audience experiences and effects. Journal of Broadcasting & Electronic Media, 35(2), 233-243. Gefen, D., & Straub, D. W. (1997). Gender Differences in the Perception and Use of E-Mail: An extension to the technology acceptance model. MIS Quarterly, 21(4), 389-400. Geisser, S. (1974). A Predictive Approach to the Random Effect Model. Biometrika, 61(1), 101107. Gerbner, G. (1998). Cultivation Analysis: An Overview. Mass Communication and Society, 1(34), 175-194. Gibbons, T., & Dixon, K. (2010). ‘Surf’s up!’: A call to take English soccer fan interactions on the Internet more seriously. Soccer & Society, 11(5), 599-613. Giddens, A. (1991). Modernity and Self-Identity: Self and society in the Late Modern Age. Stanford, CA: Stanford University Press. 136 Gladden, J. M., & Funk, D. C. (2002). Developing an Understanding of Brand Associations in Team Sport: Empirical evidence from consumers of professional sport. Journal of Sport Management, 16(1), 54. Graef, R., Csikszentmihalyi, M., & Gianinno, S. M. (1983). Measuring Intrinsic Motivation in Everyday Life. Leisure Studies, 2(2), 155-168. Greenwald, A. G., & Leavitt, C. (1984). Audience Involvement in Advertising: Four levels. Journal of Consumer Research, 11(1), 581-592. Grieve, P. G., & Hogg, M. A. (1999). Subjective Uncertainty and Intergroup Discrimination in the Minimal Group Situation. Personality and Social Psychology Bulletin, 25(8), 926940. Gurin, P., & Townsend, A. (1986). Properties of Gender Identity and Their Implications for Gender Consciousness. British Journal of Social Psychology, 25(2), 139-148. Gwinner, K., & Swanson, S. R. (2003). A Model of Fan Identification: Antecedents and sponsorship outcomes. The Journal of Services Marketing, 17(2/3), 275. Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2017). A Primer on Partial Least Squares Structrual Equation Modeling (PLS-SEM) (2nd ed.). Los Angeles, CA: Sage. Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory & Practice, 19(2), 139-152. Hair, J. F., Ringle, C. M., & Sarstedt, M. (2012). Partial Least Squares: The better approach to structural equation modeling? Long Range Planning, 45(5–6), 312-319. Hair, J. F., Ringle, C. M., & Sarstedt, M. (2013). Partial Least Squares Structural Equation Modeling: Rigorous applications, better results and higher acceptance. Long Range Planning, 46(1–2), 1-12. Hair, J. F., Sarstedt, M., Pieper, T. M., & Ringle, C. M. (2012). The Use of Partial Least Squares Structural Equation Modeling in Strategic Management Research: A review of past practices and recommendations for future applications. Long Range Planning, 45(5–6), 320-340. Hair, J. F., Sarstedt, M., Ringle, C. M., & Mena, J. A. (2012). An Assessment of the Use of Partial Least Squares Structural Equation Modeling in Marketing Research. Journal of the Academy of Marketing Science, 40(3), 414-433. Hall, J., O’Mahony, B., & Vieceli, J. (2010). An Empirical Model of Attendance Factors at Major Sporting Events. International Journal of Hospitality Management, 29(2), 328334. 137 Hartmann, T., & Klimmt, C. (2006). Gender and Computer Games: Exploring females’ dislikes. Journal of Computer-Mediated Communication, 11(4), 910-931. Havitz, M. E., & Dimanche, F. (1997). Leisure Involvement Revisited: Conceptual conundrums and measurement advances. Journal of Leisure Research, 29(3), 245. Havitz, M. E., & Dimanche, F. (1999). Leisure Involvement Revisited: Drive properties and paradoxes. Journal of Leisure Research, 31(2), 122. Havitz, M. E., & Howard, D. R. (1995). How Enduring Is Enduring Involvement? A seasonal examination of three recreational activities. Journal of Consumer Psychology, 4(3), 255276. Heere, B., & James, J. D. (2007a). Sports Teams and Their Communities: Examining the Influence of External Group Identities on Team Identity. Journal of Sport Management, 21(3), 319-337. Heere, B., & James, J. D. (2007b). Stepping Outside the Lines: Developing a multi-dimensional team identity scale based on social identity theory. Sport Management Review, 10(1), 6591. Heere, B., Walker, M., Yoshida, M., Ko, Y. J., Jordan, J. S., & James, J. D. (2011). Brand Community Development through Associated Communities: Grounding community measurement within social identity theory. Journal of Marketing Theory and Practice, 19(4), 407-422. Henseler, J. (2012). Why Generalized Structured Component Analysis Is Not Universally Preferable to Structural Equation Modeling. Journal of the Academy of Marketing Science, 40(3), 402-413. Henseler, J., Ringle, C., & Sarstedt, M. (2015). A New Criterion for Assessing Discriminant Validity in Variance-Based Structural Equation Modeling. Journal of the Academy of Marketing Science, 43(1), 115-135. Hill, B., & Green, B. C. (2000). Repeat Attendance as a Function of Involvement, Loyalty, and the Sportscape Across Three Football Contexts. Sport Management Review (Sport Management Association of Australia & New Zealand), 3(2), 145-162. Hirt, E. R., Zillmann, D., Erickson, G. A., & Kennedy, C. (1992). Costs and Benefits of Allegiance: Changes in fans' self-ascribed competencies after team victory versus defeat. Journal of Personality and Social Psychology, 63(5), 724-738. Hoffman, R. M. (2006). Gender Self-Definition and Gender Self-Acceptance in Women: Intersections With Feminist, Womanist, and Ethnic Identities. Journal of Counseling & Development, 84(3), 358-372. 138 Hogg, M. A., & Grieve, P. G. (1999). Social Identity Theory and the Crisis of Confidence in Social Psychology: A commentary, and some research on uncertainty reduction. Asian Journal of Social Psychology, 2(1), 79-93. Hogg, M. A., & Hains, S. C. (1996). Intergroup Relations and Group Solidarity: Effects of group identification and social beliefs on depersonalized attraction. Journal of Personality and Social Psychology, 70(2), 295-309. Hogg, M. A., & Reid, S. A. (2006). Social Identity, Self-Categorization, and the Communication of Group Norms. Communication Theory, 16(1), 7-30. Hogg, M. A., & Turner, J. C. (1985). When Liking Begets Solidarity: An experiment on the role of interpersonal attraction in psychological group formation. British Journal of Social Psychology, 24(4), 267-281. Hollebeek, L. D., Glynn, M. S., & Brodie, R. J. (2014). Consumer Brand Engagement in Social Media: Conceptualization, scale development and validation. Journal of Interactive Marketing, 28(2), 149-165. Hunt, K. A., Bristol, T., & Bashaw, R. E. (1999). A Conceptual Approach to Classifying Sports Fans. Journal of Services Marketing, 13(6), 439-452. Hutchins, B. (2011). The Acceleration of Media Sport Culture. Information, Communication & Society, 14(2), 237-257. IBM Corp. (2013). Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY. Iso-Ahola, S. E. (1989). Motivation for Leisure. In Jackson, T. B. E. (Ed.), Understanding leisure and recreation: Mapping the past, charting the future (pp. 247-279). State College, PA: Venture. Iwasaki, Y., & Havitz, M. E. (2004). Examining Relationships between Leisure Involvement, Psychological Commitment and Loyalty to a Recreation Agency. Journal of Leisure Research, 36(1), 45-72. Izzo, G. M., Munteanu, C., Langford, B. E., Ceobanu, C., Dumitru, I., & Nichifor, F. (2011). Sport Fans’ Motivations: An investigation of Romanian soccer spectators. Journal of International Business and Cultural Studies, 5, 96-107. Jain, K., & Srinivasan, N. (1990). An Empirical Assessment of Multiple Operationalizations of Involvement. Advances in Consumer Research, 17(1), 594-602. Johnson, B. T., & Eagly, A. H. (1989). Effects of Involvement on Persuasion: A meta-analysis. Psychological Bulletin, 106(2), 290-314. 139 Joreskog, K. G., & Goldberger, A. S. (1975). Estimation of a Model with Multiple Indicators and Multiple Causes of a Single Latent Variable. Journal of the American Statistical Association, 70(351), 631-639. Kahle, L. R., Kambara, K. M., & Rose, G. M. (1996). A Functional Model of Fan Attendance Motivations for College Football. Sport Marketing Quarterly, 5, 51-60. Kaiser, H., Hunka, S., & Bianchini, J. (1971). Relating Factors Between Studies Based Upon Different Individuals. Multivariate Behavioral Research, 6(4), 409-422. Kaiser, H. F. (1958). The Varimax Criterion for Analytic Rotation in Factor Analysis. Psychometrika, 23(3), 187-200. Kapferer, J. N., & Laurent, G. (1993). Further Evidence on the Consumer Involvement Profile: Five antecedents of involvement. Psychology and Marketing, 10(4), 347-355. Karasawa, M. (1991). Toward an Assessment of Social Identity: The structure of group identification and its effects on in-group evaluations. British Journal of Social Psychology, 30(4), 293-307. Kayany, J. M., & Yelsma, P. (2000). Displacement Effects of Online Media in the SocioTechnical Contexts of Households. Journal of Broadcasting & Electronic Media, 44(2), 215-229. Kerstetter, D. L., & Koovich, G. M. (1997). An Involvement Profile of Division I Women's Basketball Spectators. Journal of Sport Management, 11(3), 234. Klarner, P., Sarstedt, M., Hoeck, M., & Ringle, C. M. (2013). Disentangling the Effects of Team Competences, Team Adaptability, and Client Communication on the Performance of Management Consulting Teams. Long Range Planning, 46(3), 258-286. Knox, S., & Walker, D. (2003). Empirical Developments in the Measurement of Involvement, Brand Loyalty and Their Relationship in Grocery Markets. Journal of Strategic Marketing, 11(4), 271-286. Kokkinaki, F., & Lunt, P. (1997). The Relationship between Involvement, Attitude Accessibility and Attitude-Behaviour Consistency. British Journal of Social Psychology, 36(4), 497509. Kraszewski, J. (2008). Pittsburgh in Fort Worth: Football bars, sports television, sports fandom, and the management of home. Journal of Sport & Social Issues, 32(2), 139-157. Krier, D. H. (2016). Sports Team Identity & Sports Media Consumption Motivations as Predictors of Total Sports Media Consumption. Michigan State University, East Lansing, MI. 140 Ksiazek, T. B., Malthouse, E. C., & Webster, J. G. (2010). News-seekers and Avoiders: Exploring patterns of total news consumption across media and the relationship to civic participation. Journal of Broadcasting & Electronic Media, 54(4), 551-568. Kyle, G. T., & Mowen, A. J. (2005). An Examination of the Leisure Involvement-Agency Commitment Relationship. Journal of Leisure Research, 37(3), 342-363. L’Etang, J. (2006). Public Relations and Sport in Promotional Culture. Public Relations Review, 32(4), 386-394. Laurent, G., & Kapferer, J. N. (1985). Measuring Consumer Involvement Profiles. Journal of Marketing Research, 22(1), 41-53. Laverie, D. A., & Arnett, D. B. (2000). Factors Affecting Fan Attendance: The influence of identity salience and satisfaction. Journal of Leisure Research, 32(2), 225-246. Liu, Y., & Shrum, L. J. (2002). What is Interactivity and Is It Always such a good thing? Implications of definition, person, and situation for the influence of interactivity on advertising effectiveness. Journal of Advertising, 31(4), 53-64. Lock, D., Darcy, S., & Taylor, T. (2009). Starting with a Clean Slate: An analysis of member identification with a new sports team. Sport Management Review, 12(1), 15-25. Lock, D., Taylor, T., Funk, D., & Darcy, S. (2012). Exploring the Development of Team Identification. Journal of Sport Management, 26(4), 283-294. Lodahl, T. M., & Kejner, M. (1965). The Definition and Measurement of Job Involvement. Journal of Applied Psychology, 49(1), 24-33. Lowry, P. B., & Gaskin, J. (2014). Partial Least Squares (PLS) Structural Equation Modeling (SEM) for Building and Testing Behavioral Causal Theory: When to choose it and how to use it. Professional Communication, IEEE Transactions on, 57(2), 123-146. Luhtanen, R., & Crocker, J. (1992). A Collective Self-Esteem Scale: Self-evaluation of one's social identity. Personality and Social Psychology Bulletin, 18(3), 302-318. MacCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999). Sample Size in Factor Analysis. Psychological methods, 4(1), 84. MacKinnon, D. P., Fairchild, A. J., & Fritz, M. S. (2006). Mediation Analysis. Annual Review of Psychology, 58(1), 593-614. Madrigal, R. (1995). Cognitive and Affective Determinants of Fan Satisfaction with Sporting Event Attendance. Journal of Leisure Research, 27(3), 205-227. 141 Madrigal, R. (2000). The Influence of Social Alliances with Sports Teams on Intentions to Purchase Corporate Sponsors' Products. Journal of Advertising, 29(4), 13-24. Madrigal, R. (2001). Social Identity Effects in a Belief-Attitude-Intentions Hierarchy: Implications for corporate sponsorship. Psychology and Marketing, 18(2), 145-165. Mael, F., & Ashforth, B. E. (1992). Alumni and Their Alma Mater: A partial test of the reformulated model of organizational identification. Journal of Organizational Behavior, 13(2), 103-123. Mael, F. A., & Tetrick, L. E. (1992). Identifying Organizational Identification. Educational and Psychological Measurement, 52(4), 813-824. Mahony, D. F., Madrigal, R., & Howard, D. (2000). Using the Psychological Commitment to Team (PCT) Scale to Segment Sport Consumers Based on Loyalty. Sport Marketing Quarterly, 9(1), 15. Malthouse, E. (2001). Dealing with Outliers. Journal of Consumer Psychology, 10(1/2), 111112. Mangold, W. G., & Faulds, D. J. (2009). Social Media: The new hybrid element of the promotion mix. Business Horizons, 52(4), 357-365. Maslow, A. H. (1943). A Theory of Human Motivation. Psychological Review, 50(4), 370-396. Maslow, A. H., Frager, R., Fadiman, J., McReynolds, C., & Cox, R. (1970). Motivation and Personality. New York, NY: Harper & Row. McCarthy, B. (2012). Consuming Sports Media, Producing Sports Media: An analysis of two fan sports blogospheres. International Review for the Sociology of Sport, 48(4), 421-434. McIntyre, N. (1989). The Personal Meaning of Participation: Enduring involvement. Journal of Leisure Research, 21(2), 167. McQuarrie, E. F., & Munson, J. M. (1987). The Zaichowsky Personal Involvement Inventory: Modification and extension. Advances in Consumer Research, 14(1), 36-40. McQuarrie, E. F., & Munson, J. M. (1992). A Revised Product Involvement Inventory: Improved Usability and Validity. Advances in Consumer Research, 19(1), 108-115. Meir, R., & Scott, D. (2007). Tribalism: Definition, identification and relevance to the marketing of professional sports franchises. International Journal of Sports Marketing and Sponsorship, 8(4), 43-59. 142 Melnick, M. J., & Wann, D. L. (2011). An Examination of Sport Fandom in Australia: Socialization, team identification, and fan behavior. International Review for the Sociology of Sport, 46(4), 456-470. Miller, D. T. (1976). Ego Involvement and Attributions for Success and Failure. Journal of Personality and Social Psychology, 34(5), 901-906. Mitchell, A. A. (1979). Involvement: A potentially important mediator of consumer behavior. Advances in Consumer Research, 6(1), 191-196. Mittal, B., & Lee, M. (1989). A Causal Model of Consumer Involvement. Journal of Economic Psychology, 10(3), 363-389. Murrell, A. J., & Dietz, B. (1992). Fan Support of Sport Teams: The effect of a common group identity. Journal of Sport and Exercise Psychology, 14(1), 28-39. Myers, N. D., Ahn, S., & Jin, Y. (2011). Sample Size and Power Estimates for a Confirmatory Factor Analytic Model in Exercise and Sport: A Monte Carlo approach. Research Quarterly for Exercise & Sport, 82(3), 412-423. Ndubisi, N. O. (2006). Effect of Gender on Customer Loyalty: A relationship marketing approach. Marketing Intelligence & Planning, 24(1), 48-61. Oliver, M. B., Weaver III, J. B., & Sargent, S. L. (2000). An Examination of Factors Related to Sex Differences in Enjoyment of Sad Films. Journal of Broadcasting & Electronic Media, 44(2), 282-300. Ophir, E., Nass, C., & Wagner, A. D. (2009). Cognitive Control in Media Multitaskers. Proceedings of the National Academy of Sciences, 106(37), 15583-15587. Osborne, A. C., & Coombs, D. S. (2013). Performative Sport Fandom: An approach to retheorizing sport fans. Sport in Society, 16(5), 672-681. Osborne, J., & Waters, E. (2002). Four Assumptions of Multiple Regression that Researchers Should Always Test. Practical Assessment, Research & Evaluation, 8(2), 1-9. Pea, R., Nass, C., Meheula, L., Rance, M., Kumar, A., Bamford, H., Zhou, M. (2012). Media Use, Face-to-Face Communication, Media Multitasking, and Social Well-Being among 8- to 12-Year-Old Girls. Developmental Psychology, 48(2), 327-336. Pegoraro, A. L., Ayer, S. M., & O'Reilly, N. J. (2010). Consumer Consumption and Advertising Through Sport. American Behavioral Scientist, 53(10), 1454-1475. 143 Pelletier, L. G., Fortier, M. S., Vallerand, R. J., Tuson, K. M., Briere, N. M., & Blais, M. R. (1995). Toward a New Measure of Intrinsic Motivation, Extrinsic Motivation, and Amotivation in Sports: The Sport Motivation Scale (SMS). Journal of Sport and Exercise Psychology, 17(1), 35-35. Persaud, A., & Azhar, I. (2012). Innovative Mobile Marketing via Smartphones. Marketing Intelligence & Planning, 30(4), 418-443. Phinney, J. S. (1990). Ethnic identity in adolescents and adults: Review of research. Psychological Bulletin, 108(3), 499-514. Phinney, J. S. (1992). The Multigroup Ethnic Identity Measure: A new scale for use with diverse groups. Journal of Adolescent Research, 7(2), 156-176. Phinney, J. S. (1996). When We Talk about American Ethnic Groups, What Do We Mean? American Psychologist, 51(9), 918-927. Pilotta, J. J., Schultz, D. E., Drenik, G., & Rist, P. (2004). Simultaneous Media Usage: A critical consumer orientation to media planning. Journal of Consumer Behaviour, 3(3), 285-292. PricewaterhouseCoopers. (2015). Changing the Game: Outlook for the global sports market to 2015. Retrieved April 23, 2016, from http://www.pwc.com/gx/en/hospitalityleisure/pdf/changing-the-game-outlook-for-the-global-sports-market-to-2015.pdf Ray, M. L. (1979). Involvement and Other Variables Mediating Communication Effects as Opposed to Explaining All Consumer Behavior. Advances in Consumer Research, 6(1), 197-199. Reinartz, W., Haenlein, M., & Henseler, J. (2009). An Empirical Comparison of the Efficacy of Covariance-Based and Variance-Based SEM. International Journal of Research in Marketing, 26(4), 332-344. Rideout, V. J., Foehr, U. G., & Roberts, D. F. (2010). Generation M2: Media in the lives of 8-to 18-year-olds. Henry J. Kaiser Family Foundation. Rigdon, E. E. (2014). Rethinking Partial Least Squares Path Modeling: Breaking chains and forging ahead. Long Range Planning, 47(3), 161-167. Rigdon, E. E., Ringle, C. M., & Sarstedt, M. (2010). Structural Modeling of Heterogeneous Data with Partial Least Squares, Review of Marketing Research, (7), 255-296. Ringle, C. M., Sarstedt, M., & Straub, D. W. (2012). A Critical Look at the Use of PLS-SEM in MIS Quarterly. MIS Quarterly, 36(1), iiv-8. Ringle, C. M., Wende, S., & Becker, J.-M. (2015). “SmartPLS 3.” Bönningstedt: SmartPLS GmbH. Retrieved January 7, 2017 from http://www.smartpls.com. 144 Roberts, D. F., & Foehr, U. G. (2008). Trends in Media Use. The future of children, 18(1), 1137. Robinson, J. P., Kestnbaum, M., Neustadtl, A., & Alvarez, A. (2000). Mass Media Use and Social Life among Internet Users. Social Science Computer Review, 18(4), 490-501. Robinson, M. J., & Trail, G. T. (2005). Relationships among Spectator Gender, Motives, Points of Attachment, and Sport Preference. Journal of Sport Management, 19(1), 58-80. Roccas, S., Sagiv, L., Schwartz, S., Halevy, N., & Eidelson, R. (2008). Toward a Unifying Model of Identification with Groups: Integrating theoretical perspectives. Personality & Social Psychology Review, 12(3), 280-306. Rothschild, M. L. (1984). Perspectives on Involvement: Current problems and future directions. Advances in Consumer Research, 11(1), 216-217. Roy, D. P., & Goss, B. D. (2007). A Conceptual Framework of Influences on Fantasy Sports Consumption. Marketing Management Journal, 17(2), 96-108. Salsman, J. M., Brown, T. L., Brechting, E. H., & Carlson, C. R. (2005). The Link between Religion and Spirituality and Psychological Adjustment: The Mediating Role of Optimism and Social Support. Personality and Social Psychology Bulletin, 31(4), 522535. Sarstedt, M. (2008). A Review of Recent Approaches for Capturing Heterogeneity in Partial Least Squares Path Modelling. Journal of Modelling in Management, 3(2), 140-161. Sarstedt, M., Henseler, J., & Ringle, C. M. (2011). Multigroup Analysis in Partial Least Squares (PLS) Path Modeling: Alternative methods and empirical results, Measurement and Research Methods in International Marketing, (22), 195-218. Sarstedt, M., Ringle, C. M., & Hair, J. F. (2014). PLS-SEM: Looking back and moving forward. Long Range Planning, 47(3), 132-137. Sashi, C. M. (2012). Customer Engagement, Buyer-Seller Relationships, and Social Media. Management Decision, 50(2), 253-272. Scanlan, T. K., Simons, J. P., Carpenter, P. J., Schmidt, G. W., & Keeler, B. (1993a). An Introduction to the Sport Commitment Model. Journal of Sport & Exercise Psychology, 15(1), 1-15. Scanlan, T. K., Simons, J. P., Carpenter, P. J., Schmidt, G. W., & Keeler, B. (1993b). The Sport Commitment Model: Measurement development for the youth-sport domain. Journal of Sport & Exercise Psychology, 15(1), 16-38. 145 Schaufeli, W. B., Salanova, M., González-Romá, V., & Bakker, A. B. (2002). The Measurement of Engagement and Burnout: A two sample confirmatory factor analytic approach. Journal of Happiness Studies, 3(1), 71-92. Schmader, T., & Block, K. (2015). Engendering Identity: Toward a clearer conceptualization of gender as a social identity. Sex Roles, 73(11), 474-480. Schouten, J. W. (1991). Selves in Transition: Symbolic consumption in personal rites of passage and identity reconstruction. Journal of Consumer Research, 17(4), 412-425. Schouten, J. W., & McAlexander, J. H. (1995). Subcultures of Consumption: An ethnography of the new bikers. Journal of Consumer Research, 22(1), 43-61. Selin, S. W., & Howard, D. R. (1988). Ego Involvement and Leisure Behavior: A Conceptual Specification. Journal of Leisure Research, 20(3), 237. Seo, W. J., & Green, B. C. (2008). Development of the Motivation Scale for Sport Online Consumption. Journal of Sport Management, 22(1), 82-109. Sherif, C. W., Sherif, M., & Nebergall, R. E. (1965). Attitude and Attitude Change: The social judgment-involvement approach. Philadelphia, PA: Saunders. Sierra, J. J., Badrinarayanan, V. A., & Taute, H. A. (2015). Explaining Behavior in Brand Communities: A sequential model of attachment, tribalism, and self-esteem. Computers in Human Behavior, 55, 626-632. Skaalvik, E. M., Valåns, H., & Sletta, O. (1994). Task Involvement and Ego Involvement: Relations with academic achievement, academic self-concept and self-esteem. Scandinavian Journal of Educational Research, 38(3–4), 231-243. Sloan, L. R. (1989). The Motives of Sports Fans. In Goldstein, J. H. (Ed.), Sports, Games, and Play: Social and Psychological Viewpoints (pp. 175-240). Hillsdale, NJ: Lawrence Erlbaum Associates. Smith, R. E. (2006). Understanding Sport Behavior: A cognitive-affective processing systems approach. Journal of Applied Sport Psychology, 18(1), 1-27. Solomon, M. R. (1983). The Role of Products as Social Stimuli: A symbolic interactionism perspective. Journal of Consumer Research, 10(3), 319-329. Sports Business Daily. (2016) Sources: Fox, Big Ten closing in on media rights agreement. Retrieved April 19, 2016, from http://www.sportsbusinessdaily.com/Daily/ClosingBell/2016/04/19/Big-Ten.aspx 146 SportingNews.com. (2015). Soulful to Soulless: Former ESPN employees blame overspending, Disney for layoffs. Retrieved July 15, 2016, from http://www.sportingnews.com/othersports/news/espn-layoffs-employees-producers-disney-howie-schwab-fox-nbc-cbs-johnskipper/da6zva011tzh1je74qujuhb7n Sporting News Media. (2014). Know the Fan: The global sports media consumption report 2014 (U.S. overview). Retrieved March 22, 2016, from http://sportsvideo.org/main/files/2014/06/2014-Know-the-Fan-Study_US.pdf Statista.com. (2014a). Regional Share of Sports Media Rights Revenue from 2009 to 2012. Retrieved April 23, 2016, from http://www.statista.com/statistics/269793/distribution-ofsports-media-rights-revenue-by-region-worldwide/ Statista.com. (2014b). Which of These Methods Do You Use to Consume Sports? Retrieved April 22, 2016, from http://www.statista.com/statistics/288882/methods-sports-fans-useto-consume-sports/ Statista.com. (2015a). Licensed Sports Merchandise Wholesale Sales in the U.S. from 2008 to 2015 (in billion U.S. dollars). Retrieved April 19, 2016, from http://www.statista.com/statistics/240979/licensed-sports-merchandise-wholesale-salesin-the-us/ Statista.com. (2015b). Monthly household reach of ESPN in the United States from 2011/12 to 2014/15. Retrieved April 18, 2016, from http://www.statista.com/statistics/495878/espnreach-usa/ Statista.com. (2015c). Number of smartphone users in the U.S. from 2010 to 2018. Retrieved April 18, 2016, from http://www.statista.com/statistics/201182/forecast-ofsmartphone-users-in-the-us/ Statista.com. (2016a). Cable TV networks: Number of TV viewers of ESPN within the last 7 days in the United States (USA) from spring 2008 to spring 2015 (in millions). Retrieved April 18, 2016, from http://www.statista.com/statistics/228943/cable-tv-networks-espnwatched-within-the-last-7-days-usa/ Statista.com. (2016b). Global Revenue Earned from Sports Media Rights from 2006 to 2015 (in billion U.S. dollars). Retrieved April 23, 2016, from http://www.statista.com/statistics/269795/revenue-from-media-rights-in-sportsworldwide/ Statista.com. (2016c). Global Sports Market: Total Revenue from 2005 to 2017 (in billion U.S. dollars). Retrieved April 23, 2016, from http://www.statista.com/statistics/370560/worldwide-sports-market-revenue/ 147 Statista.com. (2016d). Most valuable media brands worldwide in 2016 (in billion U.S. dollars). Retrieved April 21, 2016, from http://www.statista.com/statistics/350663/most-valuablemedia-brands/ Statista.com. (2016e). Sports Media Rights Market Size in North America from 2006 to 2019 (in billion U.S. dollars). Retrieved April 4, 2016, from http://www.statista.com/statistics/194225/sports-media-rights-revenue-in-north-america/ Statista.com. (2016f). Walt Disney Company's Net Income from 1st Quarter 2010 to 1st Quarter 2016 (in million U.S. dollars). Retrieved April 21, 2016, from http://www.statista.com/statistics/224415/quarterly-net-income-of-the-walt-disneycompany/ Staurowsky, E. J. (2016). Women as Consumers and Fans. In Staurowsky, E. J. (Ed.), Women and Sport: Continuing a Journey of Liberation and Celebration. Champaign, IL: Human Kinetics. Stevens, J. P. (1984). Outliers and Influential Data Points in Regression Analysis. Psychological Bulletin, 95(2), 334. Stewart, B., Smith, A. C. T., & Nicholson, M. (2003). Sport Consumer Typologies: A critical review. Sport Marketing Quarterly, 12(4), 206-216. Stone, M. (1974). Cross-validatory Choice and Assessment of Statistical Predictions. Journal of the royal statistical society. Series B (Methodological), 111-147. Strömbäck, J., Djerf-Pierre, M., & Shehata, A. (2012). The Dynamics of Political Interest and News Media Consumption: A longitudinal perspective. International Journal of Public Opinion Research, 25(4), 414-435. Tajfel, H. (1970). Experiments in Intergroup Discrimination. Scientific American, 223(5), 96102. Tajfel, H. (1974). Social Identity and Intergroup Behaviour. Social Science Information, 13(2), 65-93. Tajfel, H. (1978). Social Categorization, Social Identity and Social Comparison. In Tajfel, H. (Ed.), Differentiation between social groups: Studies in the social psychology of intergroup relations (pp. 61-76). London: Academic Press. Tajfel, H. (1982). Social Psychology of Intergroup Relations. Annual Review of Psychology, 33(1), 1-39. Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. (1971). Social Categorization and Intergroup Behaviour. European Journal of Social Psychology, 1(2), 149-178. 148 Tajfel, H., & Turner, J. C. (1979). An Integrative Theory of Intergroup Conflict. The Social Psychology of Intergroup Relations, 3, 33-47. Taneja, H., Webster, J. G., Malthouse, E. C., & Ksiazek, T. B. (2012). Media Consumption across Platforms: Identifying user-defined repertoires. New Media & Society, 14(6), 951968. Tapp, A., & Clowes, J. (2002). From “Carefree Casuals” to “Professional Wanderers”: Segmentation possibilities for football supporters. European Journal of Marketing, 36(11/12), 1248-1269. Theokas, C. (2009). Youth Sport Participation - A view of the issues: Introduction to the special section. Developmental Psychology, 45(2), 303-306. Trail, G. T., Fink, J. S., & Anderson, D. F. (2003). Sport Spectator Consumption Behavior. Sport Marketing Quarterly, 12(1), 8-19. Trail, G. T., & James, J. D. (2001). The Motivation Scale for Sport Consumption: Assessment of the scale's psychometric properties. Journal of Sport Behavior, 24(1), 108-127. Traylor, M. B., & Joseph, W. B. (1984). Measuring Consumer Involvement in Products: Developing a general scale. Psychology & Marketing, 1(2), 65-77. Tsiotsou, R. H. (2013). Sport Team Loyalty: Integrating relationship marketing and a hierarchy of effects. Journal of Services Marketing, 27(6), 458-471. TV Insider. (2016) Most Watched Television Networks: Ranking 2015's winners and losers. Retrieved April 19, 2016, from http://www.tvinsider.com/article/62572/most-watched-tvnetworks-2015/ Underwood, R., Bond, E., & Baer, R. (2001). Building Service Brands via Social Identity: Lessons from the sports marketplace. Journal of Marketing Theory and Practice, 9(1), 113. Vinzi, V. E., Chin, W. W., Henseler, J., & Wang, H. (Eds.). (2010). Handbook of Partial Least Squares: Concepts, methods and applications. New York, NY: Springer. Vlachopoulos, S. P., Karageorghis, C. I., & Terry, P. C. (2000). Motivation Profiles in Sport: A self-determination theory perspective. Research Quarterly for Exercise and Sport, 71(4), 387-397. Voorveld, H. A. M., & van der Goot, M. (2013). Age Differences in Media Multitasking: A Diary Study. Journal of Broadcasting & Electronic Media, 57(3), 392-408. 149 Walker, M., & Kent, A. (2009). Do Fans Care? Assessing the Influence of Corporate Social Responsibility on Consumer Attitudes in the Sport Industry. Journal of Sport Management, 23(6), 743-769. Wann, D. L. (1995). Preliminary Validation of the Sport Fan Motivation Scale. Journal of Sport & Social Issues, 19(4), 377-396. Wann, D. L. (2006). Understanding the Positive Social Psychological Benefits of Sport Team Identification: The team identification-social psychological health model. Group Dynamics: Theory, Research, and Practice, 10(4), 272-296. Wann, D. L., & Branscombe, N. R. (1990). Die-Hard and Fair-Weather Fans: Effects of identification on BIRGing and CORFing tendencies. Journal of Sport & Social Issues, 14(2), 103-117. Wann, D. L., & Branscombe, N. R. (1993). Sports Fans: Measuring degree of identification with their team. International Journal of Sport Psychology, 24(1), 1-17. Wann, D. L., & Dolan, T. J. (1994). Influence of Spectators' Identification on Evaluation of the Past, Present, and Future Performance of a Sports Team. Perceptual and Motor Skills, 78(2), 547-552. Wann, D. L., Dolan, T. J., McGeorge, K. K., & Allison, J. A. (1994). Relationships between Spectator Identification and Spectators' Perceptions of Influence, Spectators' Emotions, and Competition Outcome. Journal of Sport & Exercise Psychology, 16(4), 347-364. Wann, D. L., & Grieve, F. G. (2005). Biased Evaluations of In-Group and Out-Group Spectator Behavior at Sporting Events: The importance of team identification and threats to social identity. The Journal of Social Psychology, 145(5), 531-546. Wann, D. L., Grieve, F. G., Zapalac, R. K., & Pease, D. G. (2008). Motivational Profiles of Sport Fans of Different Sports. Sport Marketing Quarterly, 17(1), 6-19. Wann, D. L., Metcalf, L. A., Adcock, M. L., Choi, C., Dallas, M. B., & Slaton, E. (1997). Language of Sports Fans: Sportugese revisited. Perceptual and Motor Skills, 85(3), 11071110. Wann, D. L., & Pierce, S. (2003). Measuring Sport Team Identification and Commitment: An empirical comparison of the sport spectator identification scale and the psychological commitment to team scale. North American Journal of Psychology, 5(3), 365-372. Wann, D. L., Raney, A. A., & Bryant, J. (Eds.) (2006). The Causes and Consequences of Sport Team Identification. London: Routledge. Wann, D. L., & Schrader, M. P. (1997). Team Identification and the Enjoyment of Watching a Sporting Event. Perceptual and Motor Skills, 84(3), 954-954. 150 Wenner, L. A., & Gantz, W. (1998). Watching Sports on Television: Audience experience, gender, fanship, and marriage. In Wenner, L. A. (Ed.), MediaSport (pp. 233-251). New York, NY: Routledge. Wetzels, M., Odekerken-Schröder, G., & van Oppen, C. (2009). Using PLS Path Modeling for Assessing Hierarchical Construct Models: Guidelines and Empirical Illustration. MIS Quarterly, 33(1), 177-195. Williams, J. (2007). Rethinking Sports Fandom: The case of European soccer. Leisure Studies, 26(2), 127-146. Witter, R. A., Stock, W. A., Okun, M. A., & Haring, M. J. (1985). Religion and Subjective WellBeing in Adulthood: A Quantitative Synthesis. Review of Religious Research, 26(4), 332342. Wold, H. (1974). Causal Flows with Latent Variables. European Economic Review, 5(1), 67-86. Wood, W., & Eagly, A. H. (2015). Two Traditions of Research on Gender Identity. Sex Roles, 73(11), 461-473. Yuan, E. (2011). News Consumption across Multiple Platforms. Information, Communication & Society, 14(7), 998-1016. Zaff, J. F., Moore, K. A., Papillo, A. R., & Williams, S. (2003). Implications of Extracurricular Activity Participation during Adolescence on Positive Outcomes. Journal of Adolescent Research, 18(6), 599-630. Zaichkowsky, J. L. (1985). Measuring the Involvement Construct. Journal of Consumer Research, 12(3), 341-352. Zaichkowsky, J. L. (1986). Conceptualizing Involvement. Journal of Advertising, 15(2), 4-34. Zaichkowsky, J. L. (1994). The Personal Involvement Inventory: Reduction, Revision, and Application to Advertising. Journal of Advertising, 23(4), 59-70. Zarrett, N., Fay, K., Li, Y., Carrano, J., Phelps, E., & Lerner, R. M. (2009). More than Child's Play: Variable- and pattern-centered approaches for examining effects of sports participation on youth development. Developmental Psychology, 45(2), 368-382. Zillmann, D., & Bryant, J. (1988). Effects of Prolonged Consumption of Pornography on Family Values. Journal of Family Issues, 9(4), 518-544. 151